Home

How to Evaluate Your Antenna Tuner Part 1

1. ES xa Op o2 be c Dd A SN s A 4 Un palanced 4 A J ke ME IL f 77 P M Cow d AR O Ite S CY 4 AS Outputs d 7 N gt Ka Je Sad a e pm M Balanced Jul Y Es ay F SN x a gt i Jutpu SN FT Sa UY No WP ke y y N n A 24 X ESS a a d fe Ny EO a K 7 ofOUNC x NW Iud ee E ES e y F Neu y OS amp SH FEN d d y y A S S Ys 3 gt Hc 7 iced J 7 deit Cel NS yl A A A DAIANCEC y s Sa i y a v Output y LDUL d Ses Zoe y A Figure 3 Circuit diagram of Johnson Matchbox antenna tuners C1 is a split stator capacitor and C2 is a dual differen tial capacitor The top unbalanced output connection is used for high impedance unbalanced loads and the other is used for low impedance unbalanced loads In the latter case the unused balanced load connection is grounded Page 7 Copyright 1996 American Radio Relay League Inc All rights reserved May 1995 QST Volume 79 Number 5 15 uH EN V YN Y e E ai add a ven ka en nm lt lt SE we Unbalanced Se NS ES NC 8 N 4 4 Outputs A wees P i ge 25 d d d A E 900r ES P lt Ground Figure 4 Circuit diagram of the Collins Model 180S 1 antenna tuner Three unbalanced configurations are available two of which form an L network and the other is a n network The tuning range
2. At 1560 W Cin L Cout Unloaded Q 1666 600 1660 Reactance 1808 044 f 603 094 AbS 699 4 Peak Unltage 14005 U 14010 Y 13465 U RAS Current 5 5 A 16 4 A 11 8 A Power Diss 54 H B14 Y 164 H k 48 1 pF 1880 8 pF k F 4 50 A 50 8 i 52 5 pH j 1673 00 E k as F Impedance 2 Series C 5 Qu 0 Hetuork MH Hain Menu CH Figure A Sample TL Screen Page 8 Copyright 1996 American Radio Relay League Inc All rights reserved April 1995 QST Volume 79 Number 4 SS 1 Pole GOTO Shac a O7 CR ry O ON Di TZ 3 uj O EN FN DN C XA ES FON CY oN y O a Iw ay Iw NA Iu oy Iw uw Ke wy A So gt gt 2 P Y A gt Es KS gt x lt lt lt S lt lt l Se lt lt ES A SE E T lt a a ZE ZE A SS SS A ZS ZE 3 C R AROUSA CO o Ara e g uc FTAA ROAR AAS INA AE short gt 6 2 6 2 6 2 124 249 49 9 100 200 402 806 162 A 4 OF NA CN Q V Ww eta Aur Lk J je xm m 8 x Larbon Ye d ei Y O WK LU 2 Figure 1A Circuit diagram for unbalanced switched resistor network Page 9 Copyright 1996 American Radio Relay League Inc All rights reserved April 1995 QST Volume 79 Number 4 to Left Balanced 2 Pole 12 Position Rotary Switch Mouser 10WR212 Output S2A O O O i O O OQ O 1 T Q T Q S1 Q Pd OX lt to O NE e Ground O e Y Mouser O Me i i 10WW034 i oe XL ide a
3. one l 8 segment and the load box eight load points are available for each SWR circle Using the 1 16 segment in tandem with the I 8 segment provides a 31 16 line The loss measurement includes the loss in these added cable segments which is small because of their short length made up a 1 16 segment and a 1 8 segment for 3 5 MHz The results for the Collins Model 180S 1 were recorded on a Smith Chart shown in Figure 5 50 Q Smith Chart paper is available from ARRL This load print a name inspired by its likeness to a footprint graphically shows the limits of the tuning range for this tuner Notice the hole in the coverage for 80 meters between 2 and 3 o clock on the chart Also loads with real parts less than 12 Q cannot be tuned with either connection Furthermore R L loads cannot be converted to 50 Q with the z connection The same method for finding loss and SWR bandwidth may be applied for complex impedance loading of the tuner The results in this case are for the tuner plus any added cable between the tuner and the resistive load box See the Appendix in Part 1 of this article The loss of the line is small and easily calculated since we know its length matched loss and the SWR present on the line so the technique of halving and doubling the load resistance will yield a good approximation for the tuner loss As we have seen characterization with the Geometric Resistance Box alone provides an enormous amount of data The
4. case the SWR is measured in the vicinity of a 2 1 SWR where the meter is more accurate Here the matched loss would be calculated using Eq 5 in Part 1 of this article Summary These days when most elements of our hobby are expensive and out of reach of many it s nice to come across a project that doesn t break the bank have presented some simple techniques for characterizing the performance of our antenna tuners We no longer have to depend on blind faith to know the influence of our antenna tuners on our antenna systems This new application for low power SWR testers is a demanding one since the accuracy must be excellent for valid results Perhaps we will see even more accurate SWR testers in the future and maybe antenna tuner manufacturers will also be inspired to improve their designs want to thank Bob Wendt KD2WN Andy Griffith W4ULD Warren Bruene W5OLY and Dean Straw N6BV for their helpful suggestions Notes 1 Frank Witt AI1H How to Evaluate Your Antenna Tuner Part i QST Apr 1995 pp 30 34 2 Autek Research Model RF 1 RF Analyst Autek Research PO Box 8772 Madeira Beach FL 33738 813 886 9515 3 These forms are available directly from the author Send an SASE for further information 4 4 M Walt Maxwell W2DU Reflections Transmission Lines and Antennas Newington ARRL 1990 pp 14 12 to 14 13 5 50 Q Smith Chart paper is available from ARRL Order Product No 1341 2 plus shipping
5. box Power Lost 10 SWR Bandwidth gt 555 Output Balance gt 67 Table 4 Johnson Kilowatt Matchbox Balanced Case Page 14 Copyright 1996 American Radio Relay League Inc All rights reserved May 1995 QST Volume 79 Number 5 Table 5 Collins Model 1805 1 Connection SWR Load 2 Band meters 160 B 40 30 20 17 15 12 10 sna Bare psf e BRE a Wa Pe Pee E Swm E S S Sws SWR BW AA Saws SWR Bu pL me SWR BV Pe ee sae Um le SWR BVa Sewa TTT CO SWR BWH Shaded boxes SWR lt 1 1 1 not possible Empty box all data within limits Blank entry in box Power Lost lt 10 SWR Bandwidth gt 5 Table 5 Collins Model 180S 1 Pi Connection Page 15 Copyright 1996 American Radio Relay League Inc All rights reserved
6. example Suppose this tuner is being used to transform a 12 5 Q resistive load on 40 meters to 50 Q From the table 32 of the power is lost in the tuner which means that 68 of the power flows on to the feed line and antenna If the input power is 1 kW then 320 W are dissipated in the tuner The lower number in that block 2 is the SWR bandwidth expressed as a percentage of the band center frequency or 144 kHz using Eq 1 The lower this number the more touchy will be the tuning of the tuner What a disquieting set of data Although tuning is achieved over a wide range many antenna loads cause very high losses in the tuner High power operation with low impedance loads could damage this tuner have successfully achieved a 1 1 SWR at the input of this tuner when the output terminals are shorted Under this condition all the transmitter s output energy is dissipated in the tuner Note also the low SWR bandwidth for many of the entries in Table 1 This means that the tuning is often difficult with this tuner The measurements were repeated with balanced loads and with the built in 4 1 balun connected These results are summarized in Table 2 The results were essentially the same as the unbalanced case except that the loss is higher because of balun loss Note that all blocks on 80 meters are filled in The parasitic effects of the internal balun were sufficient to make tuning possible Just how good a job does the balun do at providing the unbalanc
7. high and low resistance loads will exercise tuners to define some of their practical operating limits However the method describe here may be used when a complete characterization is desired Testing the Accuracy of Low Power SWR Testers For the techniques presented here to yield valid and therefore useful results the low power SWR tester must have adequate accuracy Until recently the data present here could have been obtained only with the aid of very expensive professional grade test instruments thought of the approach over three years ago but the low cost testers available then were lacking in accuracy Now the amateur has several options available used the Autek Research Model RF 1 RF Analyst to evaluate the antenna tuners in this article To test the accuracy of the RF 1 and other low power SWR testers built a special resistance box physically similar to the unbalanced Geometric Resistance Boxes in Part 1 of this article Shown in Figure 6 are the results up to an SWR of 3 1 The dc SWRs are derived from the resistances used Values for SWR values of 2 4 1 or less are important for tuner loss measurements Note that the accuracy is good over the entire HF band for that SWR range have measured two RF 1 Page 3 Copyright O 1996 American Radio Relay League Inc All rights reserved May 1995 QST Volume 79 Number 5 testers and the results are similar You can partially check the accuracy of your low power SWR tester by
8. to a low loss transmission line of the appropriate characteristic impedance the SWR on the line may be derived from the reflection coefficient measurement Since the most frequent use of the SWR tester is for standing wave ratio measurements the tester displays SWR The AEA Antenna Analyst also shows the return loss but since not all low power SWR testers have that feature only SWR readings are processed to obtain our desired results See Note 8 for a discussion of return loss reflection coefficient and SWR The TL Program for Transmission Lines and Tuners Frank Witt has developed an elegant method to measure important antenna tuner characteristics and he achieves this without creating any more QRM on our bands bless him After much discussion with Frank volunteered to rewrite my transmission line analysis program TL short for Transmission Line to include antenna tuner losses Frank could then have more test data to validate his model A version of TL is on the disk bundled into the 17th edition of The ARRL Antenna Book and this latest update is available by modem on the ARRL Hiram BBS telephone 203 666 0578 available as TL ZIP in program area seven What thought would be a simple exercise turned out to be a lot more challenging The equations for computing various network configurations get pretty hairy when losses for all possible components are thrown into the mix too finally settled on a iterative search algorithm tha
9. using the Geometric Resistance Box at the 100 50 and 25 Q settings If you get results similar to those shown in Figure 6 over the HF band then you should have confidence that your loss tests will have some value Comments and Recommendations QRPers take note You have no power available to waste to warm up an antenna tuner The data presented here suggests that it is important to determine the antenna tuner loss thought that would be recommending a kilowatt rated tuner for QRPers just to keep the loss down However my measurements of the Johnson Matchboxes suggest that a low power unit can have low loss In any event find out what the loss is What are some other applications of the techniques described here The technique used for finding the approximate loss of the balun in the Heath Model SA 2040 tuner may be used for evaluating stand alone baluns Two others that come to mind are finding the loss of feed line cable and of low pass filters the kind used to attenuate harmonics that fall in the TV and FM broadcast bands For example it is well known that one can calculate the matched loss of a transmission line by measuring the SWR with the cable either open or short circuited at the far end This method is useful only if the SWR meter is accurate at high SWR levels or if the cable is lossy The method described here where the SWR for a 50 Q cable is measured with the cable terminated in 25 or 100 Q will usually be more accurate In this
10. way to express this loss is as a percentage of the transmitter output power P oer Calculate P os from Cher 0 Prost uc 107 Eq 2 If the tuner has low loss Les can turn out to be negative which is impossible This may occur because of inaccuracies in the SWR tester If this happens assume that L sy is O dB Let s consider an example a tuner adjusted for a 1 1 SWR when terminated in R 25 O We measure S 1 6 1 when the load is changed to R 2 12 5 Q and S 1 8 1 when the load is changed to 2xR 50 2 From Eq 1 we calculate Les 1 13 dB for the 25 Q load From Eq 2 approximately 23 of the trans mitter s output power is warming up the tuner or is warming it and being radiated from it if itis not shielded What if the input power to the tuner is the legal limit 1 5 KW Then 345 W would be dissipated within the tuner This would very likely damage the tuner One advantage of this indirect method for determining loss is that it finds the totalloss This means that besides component losses losses due to poor connections radiation and damaged parts may be detected Compare the results of your measurements with that predicted from computer analysis using the TL computer program See the sidebar If there are large differences then you may have a problem in your antenna tuner Because this measurement method requires switching to adjacent loads the maximum SWR that can be tested with the Geometric Resistance Bo
11. 2 Heath Model SA 2040 Balanced Case SWR Load OO Band meters Pwr Loss Ye SWR BP Balance Pwr Loss SWR BW Balance Pwr Loss 55 SWR BAE Balance Pwr Loss 55 SWR BAM Balance ha La acm RS en zh Pur Loss 55 SWR DAS Balance Shaded boxes SWR s 1 1 1 net possible Empty box all data within limits Bian erre poa Lon 10 SWA Bantaadin gt 5 Omm Balance 67 in EA Table 2 Heath Model SA 2040 Balanced Case Page 12 Copyright 1996 American Radio Relay League Inc All rights reserved May 1995 QST Volume 79 Number 5 Table 3 Johnson Viking Low Power Matchbox Balanced Case SWR Load 0 Band meters 160 80 40 30 20 UU 05 1 10 son a C EISEN ere Pwr Loss Reie ama gir Pwr Loss pind Bus p Pwr Loss SWR BW L Balance ooo 4 Par Loss Y 3400 SWR BW Shaded boxes SWR lt 1 1 1 not possible Empty box all data within limits Blank entry in box Power Lost 10 SWR Bandwidth gt 5 Output Balance gt 67 Table 3 Johnson Viking Low Power Matchbox Balanced Case Page 13 Copyright 1996 American Radio Relay League Inc All rights reserved May 1995 QST Volume 79 Number 5 Table 4 Johnson Kilowatt Matchbox Balanced Case Or BO DOHA Pwr Loss 95 Ob Do O Mie Balance Fwr Loss S Se c REUS Balance Shaded boxes SWR s 1 1 1 nol possible Empty box all data within limits Blank entry in
12. April 1995 QST Volume 79 Number 4 How to Evaluate Your Antenna Tuner Part 1 With some simple low power measurements you can learn a lot about your antenna tuner There may be some real surprises By Frank Witt AI1H 20 Chatham Road Andover MA 01810 Antenna tuners add a great deal of flexibility to our stations by converting the antenna system impedance to the unbalanced 50 Q load required by modern transceivers Used properly an antenna tuner allows us to use one antenna on several bands A tuner with a balanced output can convert a balanced antenna into an unbalanced 50 Q load In contrast to most ham equipment older antenna tuners are still quite useful Modern ones have some frills like built in power and SWR meters and some useful switches but antique tuners are performing well in ham shacks around the world Unfortunately we often don t know much about these very simple and useful devices An antenna tuner can usually be adjusted so that the transmitter sees 50 Q but then what do you have The specifications of most commercial antenna tuners are lacking in detail Some of the things we would like to know about our antenna tuners are as follows e Tuning range What range of impedance can the tuner convert to 50 for each band Is the range different for unbalanced and balanced loads What tuner settings are needed for various bands and loads e Loss How much loss does the antenna tuner introduce into the antenna system
13. D i O XO x O Or cQ Q to Right sop O Balanced Output Short 6 2 6 2 6 2 12 4 24 9 49 9 100 200 402 806 1620 3240 Ohms 4 Watt Fim Resistors q 5 esi 1 Metal Carbon Figure 1B Circuit diagram for balanced switched resistor network Page 10 Copyright 1996 American Radio Relay League Inc All rights reserved April 1995 QST Volume 79 Number 4 Figure 2 Photo showing outside of Geometric Resistance Boxes Figure 3 Photo showing inside of Geometric Resistance Boxes The resistor leads and connections to the load switches Page 11 Copyright 1996 American Radio Relay League Inc All rights reserved April 1995 QST Volume 79 Number 4 Figure 3 Photo showing inside of Geometric Resistance Boxes The resistor leads and connections to the load switches are short Ground connection lead length is not critical TA a C rn e PU ka SF XA o E em D n a IT A Sr ey VANE U NV Ss SO PE tA DY uu nm m el UNES Ium OON ko Z em O JTO MC Ne d c Ve M LA D A SO Nd n BOTOYr CO nek e OV Sr Ly TIRAN WAY UANL O Figure 4 Antenna tuner test arrangement Page 12 Copyright O 1996 American Radio Relay League Inc All rights reserved Al April 1995 QST Volume 79 Number 4 SS 3 X via lt ca E d LEES S 2 s lt Figure 5 A balanced load that in
14. Losses in antenna tuners have been mentioned recently in QST e SWR bandwidth How far from the frequency where SWR 1 1 can move before the SWR exceeds some limit say 1 5 1 More SWR bandwidth usually means less critical tuning e Output balance Is the balanced output truly balanced e Power limit What is the maximum power the tuner will handle for various load conditions e Harmonic attenuation What attenuation of transmitter harmonics can expect from the tuner if any Except for the last two attributes everything may be measured by using a low power SWR tester Several manufacturers sell potentially suitable stand alone SWR measuring instruments In this article will describe the method for making a comprehensive set of measurements on an antenna tuner The concept is simple A known load impedance is connected to the output terminals and the low power SWR tester is connected to the input terminals The procedures for finding tuning range SWR bandwidth and output balance are straightforward Some simple analysis is needed to find the loss Low Power SWR Testers In their simplest form low power SWR testers contain a low power RF signal source and a 50 O SWR bridge Some also contain a frequency counter or a frequency synthesizer These units operate with only a few milliwatts of power and this adds enormously to their utility Antenna tuner tests may be performed on antenna systems outside the amateur bands and only low p
15. Tuners for Wire Antennas QST Apr 1994 Technical Correspondence p 84 2 James Cain Nye Viking MB V A Antenna Tuner QST Jun 1994 Product Review pp 72 73 3 Andrew S Griffith Getting the Most Out of Your T Network Antenna Tuner QST Jan 1995 pp 44 47 4 AEA Model SWR 121 HF Antenna Analyst Advanced Electronic Applications Inc PO Box C2160 2006 196th Street SW Lynnwood WA 98036 800 432 8873 Page 6 Copyright 1996 American Radio Relay League Inc All rights reserved April 1995 QST Volume 79 Number 4 5 Autek Research Model RF 1 RF Analyst Autek Research PO Box 8772 Madeira Beach FL 33738 813 886 9515 6 MFJ Model MFJ 249 HF VHF SWR Analyzer MFJ Enterprises PO Box 494 Mississippi State MS 39762 800 647 1800 Also consider MFJ Model MFJ 259 which has the features of the Model MFJ 249 plus a resistance meter 7 Kits of parts for making Geometric Resistance Boxes are available directly from the author Custom dial labels are included Send an SASE for further information 8 R Dean Straw ed The ARRL Antenna Book Newington ARRL 1994 17th ed pp 24 6 and 24 7 Where Are the Standing Waves We are using a standing wave ratio SWR tester to obtain the data but where are the standing waves There are none Actually these testers measure the magnitude of the reflection coefficient at the terminals of the network under test in our case the antenna tuner If the tester is connected
16. antenna tuners include a built in 4 1 balun for balanced loads The tuner design center for balanced loads is assumed to be 200 Q To cover the full 16 1 SWR range the range for balanced loads is from 12 5 Q 200 16 to 3200 Q 200x16 To test for output balance there is access to the center taps of the balanced loads The complete set of terminations is shown in Table 1 Notice that all load resistance values are in a geometric progression Each resistance is related by a factor of 2 or 1 2 to its adjacent neighbors The reason for this arrangement will become evident when describe the method for measuring loss Incidentally a ratio of 2 1 is not mandatory could have chosen other ratios such as 1 5 1 or 3 1 if wanted to compress or expand the test load range Geometric Resistance Boxes made two different switched resistor networks one for unbalanced load tests and one for balanced load tests The schematic diagrams for these networks are shown in Figure 1A and Figure 1B I call them Geometric Resistance Boxes because the resistance values are in a geometric progression used 1 4 W metal film 1 resistors and carbon film 5 resistors Each required resistance value was independently realized rather than trying to switch parallel or series combinations of resistors This technique keeps the parasitic inductance and capacitance the same for various switch settings Plastic boxes from Radio Shack part no 270 222 are well suited to ho
17. cludes the common mode impedance to ground Z Table 1 Loads Geometric Resistance Boxes SWR Created at Output by Load RREF 50 Q 1 5625 6 1 3 125 6 25 CA N 1 1 12 5 1 25 1 50 1 100 1 200 1 400 800 1600 Oh Se ee alae Load Resistance Balanced LoadResistance Q RREF 200 Q 6 25 wi ct 12 5 wict 25 wlct 50 w ct 100 w ct 200 w ct 400 w ct 800 w ct 1600 w ct 3200 w ct 6400 w ct Page 13 Copyright 1996 American Radio Relay League Inc All rights reserved April 1995 QST Volume 79 Number 4 These values are present primarily to make loss measurements Page 14 Copyright 1996 American Radio Relay League Inc All rights reserved May 1995 QST Volume 79 Number 5 How to Evaluate Your Antenna Tuner Part 2 Al1H shows data taken using a low power SWR tester and his simple load boxes to evaluate several antenna tuners By Frank Witt AMH 20 Chatham Road Andover MA 01810 Part 1 described how to use a low power SWR tester and a Geometric Resistance Box to evaluate antenna tuners in the unbalanced and balanced modes Results with Some Typical Antenna Tuners evaluated four antenna tuners using the procedures described in Part 1 These were selected because they represent three different design approaches They are all old and available only on the used equipment market but they are still very useful devices Some of my results may surprise you The quality of the evalua
18. d that by simply using the geometric average of the two values of reflection coefficient magnitude derived from the SWR readings for the loss calculation the result is very close to the actual loss Eq 1 in the Antenna Tuner Loss section which uses the SWR readings is equivalent to taking the geometric average of the two reflection coefficient magnitudes This avoids converting SWR to pl finding the geometric average and then using an expression for loss in terms of p This method gives the loss to within a few tenths of a dB if the SWR measurement is accurate Frank Witt was licensed in 1948 and has held the calls W3NMU K2TOP W1DTV and EISVUT He holds BS and MS degrees in Electrical Engineering from The Johns Hopkins University and is a Life Member of the IEEE Frank retired in 1991 from AT amp T Bell Laboratories after 37 years of service He is one of five hams in his family including his wife Barbara N1DIS Mike N1BMI Chris NTBDT and Jerry NTBEB Frank s novel Amateur Radio contributions include the Top Loaded Delta Loop the Coaxial Resonator Match the Transmission Line Resonator and now the Geometric Resistance Box Some of his other interests include volunteer work tennis cross country skiing and windsurfing Frank can be found in a weekly schedule on approximately 3 82 MHz on Thursdays at 9 30 PM Eastern Time Others interested in antenna system projects are welcome to join in Notes 1 John Belrose Automatic Antenna
19. e value of Z is zero Q that is when the center tap of the load is grounded The output balance test is performed as follows 1 Adjust the antenna tuner for a 1 1 SWR with the load s center tap ungrounded 2 Ground the center tap with the grounding switch on the balanced Geometric Resistance Box and observe the SWR at the input to the tuner Record the SWR which we will call S 3 Calculate output balance BAL from 100 Ss is a measure of the relative reflected power caused by the grounding of the center tap of the balanced load Since it has no precise relationship to the common mode radiation which the unbalanced generator causes derive from S amp using Eq 4 a scoring of the output balance quality of the antenna tuner If BAL 100 then we know that the output balance is perfect and the antenna tuner will not cause any undesirable feed line radiation Although it is arbitrary feel that a value of BAL lt 67 S gt 1 5 shows that the output balance of the tuner is lacking Just how detrimental balance deficiencies will be depends on the particular antenna system Page 5 Copyright 1996 American Radio Relay League Inc All rights reserved April 1995 QST Volume 79 Number 4 Results with Some Typical Antenna Tuners Looking Forward to Part 2 evaluated three antenna tuner types using the procedures described above selected them because they represent three different design approaches They are all old and availab
20. ed to balanced transformation Examine the output balance which is the lower number in the data boxes of Table 2 Clearly on 10 meters and to a lesser extent on 20 17 15 and 12 meters the balun in the Heath Model SA 2040 is lacking Although the settings of the tuner were not exactly the same for the unbalanced and balanced tests an approximate measure of the balun loss may be found by determining the difference between the two sets of measurements when the loss is expressed in decibels and calculated from Eq 1 of Part 1 This result is shown in Figure 2 The balun loss on 40 meters for all loads except 12 5 Q is quite low less than 0 8 dB However on 80 meters and to a lesser extent on other bands the balun loss is excessive for terminations away from 200 Q at the secondary Johnson Matchboxes In contrast to the other tuners described here the venerable Johnson Matchbox antenna tuner is primarily designed to provide a balanced output A low power version and a kilowatt version were sold in the 1950s and tested both models borrowing them from Tom Simmonds K1UZX and Rhyne Killian KA1CX The designs are virtually identical except for the power handling capability of the components See from the schematic of Figure 3 that except for the input link the entire circuit is balanced Table 3 displays the tuning range loss SWR bandwidth and output balance for the low power Matchbox for the balanced load case Similar data for the Kilowa
21. for five sheets See the ARRL Publications Catalog in this issue for complete ordering information 6 Mike Gruber WA1SVF MFJ 249 and MFJ 207 SWR Analyzers QST Nov 1993 Product Review pp 75 77 7 Steve Ford WB8IMY AEA SWR 121 HF Antenna Analyst QST Nov 1994 Product Review pp 77 79 8 Kits of parts for making SWR load boxes are available directly from the author Custom dial labels are included Send an SASE for further information 9 R Dean Straw N6BV ed The ARRL Antenna Book Newington ARRL 1994 17th ed p 24 23 Page 4 Copyright 1996 American Radio Relay League Inc All rights reserved Unbalanced Outpu A e LN Nu S M nre 4 N a e n 10 E amp Ut ouna AS fi 4 ral J Built In Balun May 1995 QST Volume 79 Number 5 y Se L3 we anceg DUIL LL Output Figure 1 Circuit diagram of Heath Model SA 2040 The ganged capacitor is a split stator type This design is very similar to the Ultimate Transmatch described years ago in QST Page 5 Copyright 1996 American Radio Relay League Inc All rights reserved May 1995 QST Volume 79 Number 5 H Loss N Balun Figure 2 Approximate internal balun loss for the Heath Model SA 2040 antenna tuner Page 6 Copyright 1996 American Radio Relay League Inc All rights reserved May 1995 QST Volume 79 Number 5
22. ge the frequency up and down to find the frequencies at which the SWR is 1 5 1 The difference between these frequencies is BW Do not change the settings of the antenna tuner during this test 2 express the 1 5 1 SWR bandwidth SWRBW as a percentage of the band center frequency Calculate SWRBW from SWRBW 2215 Eqs 0 where F band center frequency in the same units as BW 5 3 At the end of this test restore the test frequency to the nominal test frequency Output Balance Test A major application of an antenna tuner is for driving low loss balanced feed lines Up to this point all the properties evaluated apply to both unbalanced and balanced load applications A characteristic that applies to only a balanced load application is output balance When the load is balanced such as a center fed wire antenna fed with ladder line the generator should also be balanced Radiation from the feed line can occur if the generator is not balanced A representation of a balanced load is shown in Figure 5 The load impedance Z is split into two equal parts At the center tap there is a common mode impedance Z shown as an impedance between the center tap and ground The value of Z will vary greatly with frequency and from system to system If the generator is balanced to ground no current will flow in Ze and its value will have no practical significance When the generator is not balanced a worst case situation occurs if th
23. in the interconnecting wires or straps and contact resistance in switches and where connections are made In unshielded tuners radiation is another form of loss Eddy current losses in shield materials contribute Usually losses in insulators are small but if arcing has occurred the charred residue may introduce loss too The amount of loss depends on the frequency and the load impedance Why do we want to know the loss Losses add up in antenna systems There are situations where the tuner is more lossy than the feed line Further loss can be destructive If you measure high losses under some load conditions this means that when high power is used high heat dissipation can be concentrated in a small volume and parts can be destroyed If the antenna tuner loss is excessive get another tuner or find other approaches for feeding your antenna Of course you can directly measure antenna tuner loss with a 50 dummy load a power meter and your station transmitter With the tuner adjusted for the 50 O load the power into the load is measured with and without the tuner in the circuit The ratio of these two power readings expressed in dB is the antenna tuner loss However this method doesn t tell much about the antenna tuner why use an antenna tuner for a 50 O load Or you can indirectly determine the antenna tuner loss at low power and with different loads Refer again to Figure 4 Assume for the moment that an antenna tuner is terminated in so
24. is impressive Page 8 Copyright 1996 American Radio Relay League Inc All rights reserved May 1995 QST Volume 79 Number 5 Figure 5 Load print of the Collins Model 180S 1 antenna tuner for 3 5 MHz The Xs and Os define the region where a 1 1 SWR is achieved with the p connection and the L connection respectively The speckled region shows the amount of the impedance range expansion which the alternate L configuration provides Page 9 Copyright 1996 American Radio Relay League Inc All rights reserved May 1995 QST Volume 79 Number 5 resistive Load lt 50 ohms resistive Load lt 50 ohms 20 au 0 10 20 50 Frequency MHz Frequency MHz Figure 6 Evaluation of the Autek Research Model RF 1 RF Analyst Page 10 Copyright 1996 American Radio Relay League Inc All rights reserved May 1995 QST Volume 79 Number 5 Table 1 Heath Model SA 2040 Unbalanced Case SWR Load 0 Band meters 160 B 40 KA 20 17 15 12 10 aid AAA Bai ie dell ed De lol NA SWR 2 3 3 3 TOSS IEEE ID SWR 4 3 Ho A Dall ll el al SWR El 4 4 TH HAARAMA WARATA PIREU LIT extr Ire Shaded boxes 5WR s 1 1 1 nat possible Empty box all data within limits Blank entry in box Power Lost lt 10 SWR Bandwidth gt 5 Table 1 Heath Model SA 2040 Unbalanced Case Page 11 Copyright 1996 American Radio Relay League Inc All rights reserved May 1995 QST Volume 79 Number 5 Table
25. le only on the used equipment market but they are potentially still very useful devices Some of my results presented in Part 2 may surprise you Appendix Loss From SWR Measurements For transmission lines and matched resistive attenuators it is possible to determine the matched loss by making SWR measurements at one end and changing a resistive termination at the other end When the matched terminating load resistance is either halved or doubled the loss of the line or attenuator is given by S 1 where S the SWR at the input of the line The use of SWR measurements at the input end of an antenna tuner under test provides only an approximate value for the loss Here s why Assume that a resistor R is connected to the output of the tuner Adjust the tuner to provide a 1 1 SWR at the input Connect a 50 Q resistor to the input terminals of the tuner If the output impedance of the tuner under these conditions equals R then the SWR at the input when the load impedance is R 2 or 2xR will be the same This value of SWR which is a way of expressing the relative power reflected may be used in Eq 5 to determine the loss of the tuner In the real world this condition that the output impedance of the tuner be R is not usually true If the tuner is lossless the output impedance will be R and the measured SWR will equal 2 1 for terminations of R 2 and 2xR For a lossy tuner the two SWR readings will usually be different have discovere
26. me R say 100 O Assume also that the tuner has no loss and that it is adjusted so that the SWR is 1 1 at its input If the load resistance is halved 50 Q or doubled 200 Q without retuning the tuner the SWR at the input will also be 2 1 This comes about because no energy is dissipated in the Page 3 Copyright 1996 American Radio Relay League Inc All rights reserved April 1995 QST Volume 79 Number 4 antenna tuner The antenna tuner acts strictly as an impedance transformer in this case a lossless 2 1 impedance transformer However if there is loss in the antenna tuner the SWR at the input for a 2 1 SWR at the output will not be 2 1 This gives us an indirect way to find the antenna tuner loss More details on the theory behind this approach are in the Appendix The low power SWR tester can measure the tuner loss over a wide range of load impedances Refer again to the test setup of Figure 4 Follow these steps to obtain the loss 1 Adjust the tuner for a 1 1 SWR with a load resistance of R Note unless the SWR is at least 1 1 1 this method to find the loss will not work 2 Change the load resistance to R 2 one switch position lower on the test box Record the SWR and call it S4 3 Change the load resistance to 2xR one position higher from the starting position in 1 Record the SWR and call it S 4 Calculate an estimate of the loss L es in dB from S S 1 9 1 5 1 Loan blog Eq 1 5 Another
27. nce a 1 1 SWR is achieved Measurements are made at 1 8 3 5 and 29 7 MHz to establish the tuning range at the extreme band edges For other bands the midband frequency is used Characterization is done by using the setup of Figure 4 Refer to the later sections for detailed descriptions of each test Follow these steps 1 Set the frequency 2 Set the load resistance to 50 for unbalanced tuners or 200 Q for balanced tuners 1 1 output SWR Page 2 Copyright 1996 American Radio Relay League Inc All rights reserved April 1995 QST Volume 79 Number 4 3 Perform the Tuning Range test Adjust the tuner for 1 1 SWR I settle for a 1 1 1 SWR if 1 1 is unattainable If you cannot obtain this condition record this fact and move to an adjacent load resistance and try to adjust for a 1 1 SWR Repeat and record the settings of the tuner controls 4 Perform the Loss test 5 Perform the SWR Bandwidth test Afterwards reset the test frequency 6 Perform the Output Balance test on balanced tuners 7 Switch in another load usually an adjacent one and readjust for a 1 1 SWR Repeat Steps 3 through 6 until all loads are used 8 Change to the next test frequency and repeat Steps 2 through 7 The measurements for both balanced and unbalanced loads use the same procedure If the built in balun has a 4 1 impedance ratio the settings of the tuner should be near the appropriate ones used for the unbalanced tests For example the setting
28. ow loss for all the tested loads Also the SWR bandwidth is very large over much of the wide tuning range leading to an excellent tuning sensitivity characteristic Look at all the white space in Table 5 In spite of its age this tuner is an impressive performer It s the same size as the low power Johnson Viking Matchbox On the down side it lacks the convenient switching and metering of the modern units A More Comprehensive Characterization Technique One differentiating factor for antenna tuners is their ability to handle a wide range of loads On a Smith Chart it is easy to plot the entire range of loads that can be tuned The technique will be described for the unbalanced case have extended the method to the balanced case as well The Geometric Resistance Box adequately exercises the antenna tuner to establish many of its qualities however a complete characterization requires the use of complex impedance loads Fortunately by augmenting the resistance box with a few segments of transmission line a very complete characterization is possible The technique takes advantage of the fact that a load impedance that provides a certain SWR relative to 50 Q can be rotated on a 50 Q Smith Chart along the SWR circle One full rotation 360 is accomplished when a half wavelength of 50 Q line is inserted between the tuner and the load This means that to get 45 rotation 1 8 of a full rotation a 1 16 line is required Thus with one 1 16 segment
29. ower load components are needed A high power dummy load is not needed to evaluate high power antenna tuners The accuracy of the low power SWR tester must be high enough to obtain meaningful data We will be using data in the SWR range between 1 1 and 2 4 1 so it s important that the instrument be accurate in that range In Part 2 I ll show how to determine the accuracy Antenna Tuner Loads Page 1 Copyright 1996 American Radio Relay League Inc All rights reserved April 1995 QST Volume 79 Number 4 evaluate antenna tuners using switchable resistive loads These cover the range of impedances that might be encountered in a typical installation For unbalanced loads the tuner design center is assumed to be a 50 Q resistive termination If the load at the output of the tuner is 50 Q then a 2 1 SWR load is either 25 Q or 100 Q a 4 1 SWR load is either 12 5 Q or 200 Q etc Since I ve seen some antenna tuners specified to tune loads with SWR up to 10 1 decided to test with SWR up to 16 1 that is resistive loads between 3 125 Q 50 16 and 800 Q 50x16 The loads in my test boxes are pure resistances even though a more comprehensive test would include complex impe dance loads For example a load of 40 80 Q would also yield a 2 1 SWR For most situations however an antenna tuner may be satisfactorily characterized using only resistive loads A method for doing a more comprehensive test will be covered in Part 2 Most modern
30. s for a 100 load for the unbalanced tuner should be about the same as the settings for a 400 O load for the balanced case Make sure the ground lead of the balanced resistor network is connected to the ground terminal of the antenna tuner Tuning Range Test The useful tuning range of an antenna tuner depends on its design Although tuners are useful even if they cannot be tuned for a perfect 1 1 SWR I define the tuning range to include only impedances where the SWR can be reduced to 1 1 1 or lower Sometimes a range of tuner settings will yield a 1 1 SWR so it is important that the tuner be adjusted using the instructions found in the user s manual You will probably find that there are certain loads for which a 1 1 SWR cannot be obtained as you try various frequencies If the SWR tester also measures impedance adjust the tuner so Z 50 Q when the SWR is 1 1 This avoids the uncertainty and inaccuracy that results from a broad SWR reading While testing the tuning range with the balanced resistor network the ground lead is connected to the antenna tuner ground terminal but the center tap of the load is not grounded By the way record the control settings for future use they will serve as coarse tuning settings when you tune up with an actual antenna Antenna Tuner Loss Test Antenna tuners are made up of inductors capacitors connectors and balun transformers all of which should have relatively low losses In addition there are losses
31. t efficiently does all the required math TL analyzes four types of network configurations low pass and high pass L networks Pi networks and T networks It does not analyze link coupled circuits like the Johnson Matchbox Figure A shows a TL screen analyzing a T network tuner feeding a 100 foot long center fed dipole 50 feet high through a 100 foot length of 450 Q window line at 1 83 MHz This antenna and feed line are discussed in the 1995 ARRL Handbook Admittedly this is a pretty extreme example of how lossy an antenna tuner can be for 1500 W of power going into the tuner 972 W ends up as heat This is certainly going to create some problems not to mention the 14 000 V floating around inside the cabinet Obviously this is not an optimum solution but it illustrates the use of TL pretty dramatically Although a balun would normally be used in this application did not model this component explicitly The balun would add even more to the losses N6BV Page 7 Copyright 1996 American Radio Relay League Inc All rights reserved April 1995 QST Volume 79 Number 4 TL Ver 2 06 Copyright 1993 1995 ARAL by H6BU Frequency 1 838 MHz Transmission line 4508 0hm Window Ladder Line Length 180 00 Ft At Antenna Tuner output 4 40 j 34 93 f 35 21 at 82 82 Highest network effective Q 171 6 Estimated power lost in tuner for 1500 W 372 W t 4 54 dB 64 81 lost Transmission line loss 11 59 dB Total loss 16 12 dB
32. tion depends on the accuracy of the low power SWR tester A property of many SWR testers is that their accuracy is best at values of SWR under 3 1 All units were tested with the Autek Research Model RF 1 RF Analyst which found adequate for the task My load boxes were described in Part 1 They are themselves almost frequency independent over the HF range The layout minimizes parasitic inductance and capacitance but the leads particularly in the balanced output evaluation do cause some departure from a purely resistive load on the higher frequency bands Fortunately these parasitic effects can be regarded as a part of the antenna tuner s own parasitic effects and they are tuned out by the tuner when the tuner is used to transform a load resistance R to 50 Q Since the tuner settings are not changed when the load resistance is changed to R 2 and 2xR the compensation by the tuner remains This effect makes the load resistances very accurate for this application Also the length of the ground lead for the output balance test is not critical Recording and Displaying the Data The complete characterization of an antenna tuner involves a lot of data To meet your own needs you may wish to take only some of the data The forms use are set up to accept all the data and calculated results for a given tuner Different forms are used for the unbalanced and balanced load tests Presentation of the data is a challenge because there are nine HF amate
33. tt Matchbox is given in Table 4 Even though the units were intended to work on only 80 40 20 15 and 10 meters they cover all the newer bands except 30 meters did not evaluate these antenna tuners with unbalanced loads As can be seen from Tables 3 and 4 the loss is quite respectable Oddly enough the loss of the low power Matchbox is less than that of the high power unit The SWR bandwidth is small for many loads so the tuning tends to be touchy Comparison of Tables 3 and 4 shows that the tuning of the Kilowatt Matchbox is more sensitive but it does exhibit a wider tuning range Output balance for these tuners is the best have seen with excellent balance for all settings shown Collins Model 180S 1 Figure 4 shows the schematic of the Collins Model 180S 1 antenna tuner made from the mid 1950s through the late 1960s It is rated at 1 kW The 500 pF output capacitor is a vacuum variable unit The tuner may be connected as anL Page 2 Copyright 1996 American Radio Relay League Inc All rights reserved May 1995 QST Volume 79 Number 5 network or as a m network Only an unbalanced output connection is provided The tuning range loss and SWR bandwidth for the x connection are shown in Table 5 A similar table may be created for the optional L connection have used this tuner with an inverted L antenna on 160 meters even though that band is outside the specified range In contrast to the Heath kilowatt tuner this tuner has very l
34. ur bands and 11 loads for each band On top of this for each band and load there are several tests performed A concise and understandable display of the results is possible through the use of tables A cluttered presentation is avoided by showing data only when the loss SWR bandwidth or output balance is outside specified limits show the following explicitly in the performance tables e Power lost for Pos gt 10 e SWR bandwidth for SWRBW 5 lt 5 e Output balance for BAL 67 Admittedly these are somewhat arbitrary limits but performance within these values should prove to be satisfactory for most applications If an SWR equal to or less than 1 1 1 cannot be achieved the data box is shaded To find the 1 5 1 SWR bandwidth in kHz from the measured data in the tables use BW 10F x SWRBW Eq 1 where F is the test frequency in MHz Heath Model SA 2040 Page 1 Copyright 1996 American Radio Relay League Inc All rights reserved May 1995 QST Volume 79 Number 5 The schematic diagram of the late 1970s Heath Model SA 2040 antenna tuner is shown in Figure 1 Note that its topology is similar to that of the original Ultimate Transmatch with a split stator capacitor on the transmitter side Later versions of this tuner have used a simple T topology with some advantages The power rating of the tuner is 1 kW CW and 2 kW PEP SSB The tuning range loss and SWR bandwidth are shown in Table 1 Let s look at a specific
35. use the switches and resistors Photos of the Geometric Resistance Boxes are shown in Figures 2 and 3 Layout is critical in order to have the load resistance come close to the dc value from 1 8 to 30 MHz An SO 239 connector is used for the unbalanced resistor network Some of the ribs inside the box must be removed by cutting or grinding to properly seat the connector Connection to the SO 239 connector of the tuner is made with a double PL 259 adapter Amphenol 83 877 or Radio Shack 278 192 This arrangement provides short leads and the box is supported by the connectors In the balanced Geometric Resistance Box a ring of 20 tinned copper wire is used as the common center tap of each load A second rotary switch is used to ground the center tap during the output balance test The same switch allows grounding either of the balanced output terminals for input output isolation tests on stand alone current baluns Short clip leads are used to connect the balanced resistor network to the output terminals of the antenna tuner Make these leads just long enough so that the box can be supported by a stack of books piled behind the antenna tuner The ground lead should be made longer so it can reach the ground terminal on the tuner Stay away from the balanced output terminals when the antenna tuner is adjusted and when the SWR readings are taken Taking the Data The most time consuming task is adjusting the antenna tuner Therefore all data should be taken o
36. xes described is 16 1 This is a sufficient range for most antenna tuners used in practical applications Frequency accuracy is not critical because the loss does not change much over a small frequency range How accurate is this method The accuracy is almost totally dependent on the accuracy of the SWR tester Through computer simulation of many antenna tuners using a wide range of loads have found that this method of estimating loss is accurate to within a few tenths of a dB assuming of course that the SWR tester is perfect Fortunately some SWR testers on the market have sufficient accuracy to make this method very practical and useful SWR Bandwidth Test The SWR bandwidth gives us an indication of how difficult it is to adjust the tuner The larger the SWR bandwidth Page 4 Copyright O 1996 American Radio Relay League Inc All rights reserved April 1995 QST Volume 79 Number 4 the easier will be the tuning Very often we wish to QSY after the antenna tuner has been tuned at a particular frequency How much can we change the operating frequency and still keep the transmitter happy say by keeping the SWR below 1 5 1 If the impedance of the antenna doesn t change rapidly with frequency the antenna tuner will control the 1 5 1 SWR bandwidth seen by the transmitter Again the low power SWR tester provides a simple way to measure the SWR bandwidth Follow these steps 1 With a 1 1 SWR at the desired frequency and load resistance chan

How to Evaluate Your Antenna Tuner Part 1

Contents

Download Pdf Manuals

Related Search

Related Contents