1296 MHz Transverter

5/23/2018 1296 MHz Transverter

1/8

1296-MHz transverterComplete construction details

for a simple,inexpensive transverter

for ssb and CWthat will makea noticeable denton the 1296-MHz

amateur bandecent issues of h m r dio and other amateur

publications have contained a wealth of constructionarticles on equipment for 1296 MHz, indicating thegrowth of interest and activity on that band. Con-spicuously absent from the literature, however, is asimple, inexpensive way of generating reasonableamounts of stable transmitter output powergreater than 1 watt or serious CW and ssb workon 1296 MHz.

Most long-haul, narrow-band DX work is con-ducted at or just above 1296 MHz, leaving the lowerpart of the band for wideband modes. Traditionaltransmitting schemes for this band usually involvetripling from 432 MHz using planar triodes in a cavityor stripline arrangements, and more recently, varac-tor diodes. These approaches yield CW or fm signals,but they are obviously unsuitable or single sideband.

Recent solid-state mixer designs, although suitablefor producing clean ssb and CW signals have oneprincipal drawback: low power output, typically inthe dozens of milliwatts range. This requires con-siderable linear amplification to approach reasonablepower levels. The circuitry associated with uhf mix-

ers is difficult for amateurs unfamiliar with thesedevices and expensive in terms of dollars investedper watt of outpuut power obtained.

In an effort to overcome these drawbacks withminimum circuit complexity and financial invest-ment, a high-level mixer was developed which usesthe popular 2C391728913CXlOOA5 family of planartriodes. These tubes are abundantly available surplusat extremely reasonable prices. Depending on theplate voltage applied to the tube, this transverter willdeliver from 5 to 15 watts of clean, stable CW andssb power output on 1296 MHz. This is more thanenough for routine contacts up to a 100 miles(160km) or more. My 1296-MHz signals are regularlycopied at +20 dB over S-9 over a 50 mile (80km)path using the transverter stage alone; for moreserious DX work the unit will drive a single 7289 to100 watts ssb and CW outputtheory of operation

The 1296-MHz transverter operates like a receivingconverter or mixer in reverse, and at much higherpower levels. As shown in fig. 1, an ssb signal fromthe output of a high-frequency or vhf transmitter(here considered to be the intermedi te frequency ori- f) s mixed with a higher frequency carrier (the localoscillator or LO to produce sum and difference fre-quencies, of which one is the desired uhf ssb signal.The remaining, undesired signals are eliminated wi tha selective filter.I used the output of a %-MHz ssb transceiver formy i- f. Obviously, other ssb source frequencies couldbe used, but t is desirable to use as high an i-f aspossible to separate the desired mixer product fromthe unwanted LO and difference frequencies asmuch as possible, making t easier to eliminate theunwanted signals by filtering. Intermediate frequen-cies as low as 21 or 28 MHz can be used with little dif-ficulty.

Transceivers in the 10-20 watt class are ideal fordriving this transverter. They should, however, be

y Joe M. Cadwallader KGZMW Star Route2 Bosse Road Jackson California 95642


2/8

isolated from the transverter by a simple 3 dB at- Ftenuatorl (such as a suitable length of RG-58lUcoaxial cable) to make sure the transceiver is ter-minated in a matched, resistive load. The output oftransceivers in the 100-watt class should be at- C87289 RFC 3tenuated down to about 10 watts; don't just turndown the DRIVE or MIC GAIN control.The transverter requires about 5 watts of localoscillator injection at 1296 MHz plus or minus the i-f . I-F INPUT 63VACThis signal can be derived in a number of ways. In mycase, withh a 50-MHz i-f, a LO of either 1246 or 1346MHz was needed. A crystal-controlled signal source LOINPUTproviding about 10 watts output at 415.333 MHz wasbuilt; this was used to drive a tripler stage to 1246MHz with about 5 watts output.

4 X 4 X 2 I l O X l o X 5 m ~ OX i C1, C3 part of cathode circuit (see ig. 5)15 pF silver-mica capacitor for 50-MHz i-f (three 47-pF dippedsilver-m~ca apacitors in parallellplate tuning capacitor (see ig. 7)part of L5 (see ig. 3 or 10 pF piston trimmersnon-existent; represents dc open condition of this lineconfiguration (see ext)grid bypass capacitor (see ig. 4)

C8,C9,C10 1OOO pF feedthrough capacitors. C8 must be rated for appliedB voltage

L1 part of cathode circuit (see ig. 5)L2 4 turns no. 16 (1.3mm) enamelled copper wire on 114 (6.5mmt

slug-tuned coil form (fo r50-MHz -f)L3 1 urn no. 16 (1.3mml around cold end of L2L4 plate line (see ig. 6)

6 3 v b c GRID L5 114 (6.5mm) wide copper or brass strip, about 118 (3mn-1)BIAS away from plate line (see fig. 3)fig. 1. Circuit layout for the 1296-MHz transverter. Component de- RFC 15 turns no. 16 (1.3mm) copper wire, closewound on 1/16tails ere listed under fig. 2. (1.5mm) mandrel

There are a number of ways to generate the 415-MHz signal: the easiest is to modify and retune anexisting transmitter which operates near this fre-quency. Many 432-MHz transmitters described inamateur publications can be easily retuned; tran-sistorized transmitter kits advertised in amateurpublications are very reasonably priced and shouldwork well for this purpose.

Even old commercial 450-MHz fm transmitterstrips, often available as junk, work nicely. If this ap-proach is used, however, a few precautions are inorder: turn the DEVIATION control off and, if possi-ble, remove the speech-amplifier and phase-modulator tubes; substantially increase the powersupply filtering to assure clean output with no achum which would otherwise appear on your trans-verter LO signal; voltage-regulate the oscillator andbuffer stages with Zener diodes or VR tubes to main-tain oscillator stability; and use a good quality crystalwith a low temperature coefficient in a temperature-controlled oven, or mount the crystal under the

fig. 2. Schematic diagram of the 129 MHz trensverter. The circuitlayout Is shown in fig. 1. R f power output at 1296 MHz Is 17watts.

1246MHzCARRIER

FREOUENCYTRIPLER

DRlVER415333MH2CARRIER(IOWJ

fig. 3. Block diagram of the 129 MHz trensverter system for CWand ssb operation. Although the author used a 50-MHz ssb/CWtransmitter, 21 or 28 MHz could be used with equally good results.Frequencies below 21 MHz are not recommended because of thedifficulty in separating the resulting mixer products.


3/8

chassis away from sources of heat, to reduce LOdrift.

Regardlessof which approach is used to generatethe LO signal, a small amount can alsobecoupledoffor LO injection t the receiving converter, thusreducing the total system equipment requirementand yielding true transceive operation on 1296 MHz.In my case, a small amount of 415.333-MHz energywas inductively coupled from the PA grid circuit of a10-watt transmitter strip used as the LO source andapplied to the multiplier diode of a popular trough-l ine receiving converter.2

The task of triplingup to the transverter s requiredLO injection frequency can be readily accomplishedin a varactor multiplier either commercial* orhomebrew3 or a stripline or caviry multiplierstage4 5can be uilt around a 2C3917289 triode. Ithas been suggested that cavity assemblies fromsurplus uhf equipment such as the UPX-6 wouldserve this purpose well. As is, these beautiful cavitiestune from roughly 1 0 o 1200MHz.

Tha complutm t s M H t msblCW t m n w m r . m a u n t d on p msurird charoh. The local-oseillatot chaln Is at the upper I , thevnrmctor trlplsr Is to the loft. and tho hFph+v~ltagsower supply andbloww mra at thetight.

transverter circuit is remarkably simple, using a 7289(2C393 or equivalent in the familiar grounded-gridconfiguration. As is common practice n this applica-tion. the tube grid is not actually grwnded directlybut rather is bypassed, through capacitor C7 so ?hagrid is grounded a t the signal frequency while re-maining above ground to dc This provides a conven-ient way to apply grid bias through RFC2 and C9without affecting the rt behavior of he grid circuit.tub * t. Pmslblm I-tllacal orelllator comblnmdom lor the 12 &MHttramvsner. LO fraqumnciwabove1mMHz nvmnrhm mldsband,

intermediate loear drivarf requsncy oscillator LO 3)21 MHz 1275 MHz 425 DMIMHz21 MHz 1317 MHz 439.000 MHz28 MHz 12W MHz 422.686 MHz28 MHz 1324 MHz 44t mMH z5 MHz 1 2 4 5 M H z 4 1 5 3 3 MHz5 MHz 1346 MHz W 6MMHz

1 4 MHz 1152MHz 84 MHz144 MHz 1440 MHz 480 000MHz

The grid bypass capacitor, C7, consists of a flat,Cloabup of the cmthedu compartment. The BNC connector onrho left is used for the local osclllstor input. while tha ona on thmright iu for the I - I npur. The filament and grtd voltagw arm broughtInto tho companrnnntby fad-through capaeltors.

Table 1 ists the required local-oscillator frequen-cies for various intermediate frequencies. Keep inmind that using an LO ~ b o v e 296 MHz causes inver-sion of the sideband in the transverter: for example, a1346-MHz LO minus a 50-MHzupper sideband signalequals 1296-MHz lower sideband.

Referring to the schematic diagram, fig. 2, the'The MMv-1296 ttipler available from Spectrum Internstianal, BoxW oncord. M asssehusetts01742.

concentric brassor copper plate connected by fingerstock to the tube s grid collar and insulated from thechassis with a thin mica or Teflon sheet. This type ofbypass plate is standard equipment on mili tav sur-plus vhf communications gear and can often bescavenged. The bypass plate can also be home builtby soldering finger-stock material* around an appro-priately sized hole centered on a flat brass or coppersheet If g. 41. Thin mica for the dielectric material isavailable in moat hardware or plumbingsupply storesas replacement material for gas furnace pilot-light n-spection holes. It can be easily cut with scissors or anxacto knife.

The bypass plate is insulated from its mountingInstrument Specialties Company, t i t t l e FaHs, NewJersev07424.


4/8

screws with the =me type of nylon bushings whichare used to mount and insulate power transistors.Once assembled, a typical value for this bypasscapacitor is about 100 pF; this represents about 1ohm of capacitive reactance essentiarly a deadshort at 1296 MHz, and effectively grounds thegrid at th t frequency. However, at lower frequen-cies the grid is definitely not a t I ground: a t 5 MHzthe reactance of the grid bypass is about 30 ohms,and at 28 MHz it is abaut 55 ohms. Thus, the grid canbe driven by the low-frequency ssb i-f signal whileth cathode is driven by the high frequency LOsignal: the sum and difference of these two signalsappear in the plate circuit. tip. 4. ~o m t ru c t i ~nf tha grid b m apmeltor.12

This simple approach can be applied to numerousuhf tubes in various grounded-grid configumtions,stripline or cavity, commercial or military surplus,with equally g o d esults.

The cathode circuit, fig. 5, driven at the LO f r equency of 1246 MHz in my case, consists of ashorted quarter-wave line section t l made of thinbrass or copper sheet 1J4 inch 6.5rnmIwide, wrap-ped around the tube cathode slmve and running1-314 inch 44mm) to chassis ground. The line istuned with C1 which may be a low loss, high quality

watts, these coils should be-either air-core orwound on a low-loss slug-mned ceramic coil form ofrnodemte diameter (1 4 inch I6.5mml minimum,larger preferred) using no. 14 1.6mrn) to no. 8Irnm) enameled copper wire. C2 should be a good

quality mica ar silver-mica capacitor to minimizelosses. Two or three capacitors may be paralleled tohandle the required rf current and still resonate withC2 at the i-f. The C2-l2 combination should be p e

glass or ceramic piston trimmer or, better yet, ametal tab bent up near the line, or a brass machinescrew with a brass disc about 1/2 inch (13mrn) indiameter soldered to its end similar t o C4). The LOenergy is capacitively coupled to the middle of thisline by C3 a small brass or copper tab soldered to theLO input connector and bent up near the cathodeline, Spacing can be adjusted for maximum LO drive. tm

The ssb i-f signal is coupled to the grid circuit by w 7 m -SLCCrnL3, a one-turn link wound around the cold end of UThe U C 2 circuit must resonate at the intermediatefrequency. Since this circuit will be driven with about

Th. plawllm m n d ~ o u mor the I s M H z m n w r t w . Tha utpureoupllng naruvork. E C 5 . s to thm rlght.

IUWTC2 10pF plston trirnmerwpat~tworbuilt sC4 fig.71c3 t 14 I6.5rnml widacnpper or brass drspL1 114 16.5mml widecopps-rot brass strop

Rg 6. Comhuctlm of rho u t h o d a clrcult for the 1ZSbMHzmmvmmr. Thlo drc u l t 1 I n a a l l d n n omr ll a lurnlnum mlnlboxwklch Imwmtd on thm plmtm mclolura mmeflg. 1 and the photopnph.1.

tuned to the intermediate frequency with a grid-dipper before nstallation.

The plate circuit fig. 1)consists of an open half-wave line sect ion L4, tuned to 1296 MHz bycapacitor C4. The line section is made of sheet cop-per or brass with finger stock connections to the tubeanode ring and supported by insulating columns ofTeflon Rexolite, ceramic, or other good uhf dielec


5/8

Constmct ion datails of the plate line, output network, an d mounling of rhsgrid-bypasacapscitor,capacitor t7.Note that thsmount-ing screw on the l e f t i s connsetsd t o tha grid bypass plsra for gridbias If rom the cathode encloaule. below .

tr ic. The anode f inger stock assembly can be foundin the same military surplus vhf com mun ication gearas the grid bypass assembly. Alternatively, commer-cial finger stock may be formed to the appropr iatesize and soldered around a hole in one end af the

Another v i m of tho plots line showing thw grid bypas* capacitor.C f : output link. L5-65; plnto rf choke. RFCB: mnd bypssr capacitor.C8.

plate line, L4 large enough to accommodate thetube anode r ing as shown i n fig. 6.Capacitor C6 does not really exist, but merelyrepresents the dc open condition of this l ine con-f iguration. In fact, the entire l ine and tube anodehave B + appl ied through RFC3 and C8, This feed-through bypass capacitor must be rated t o wi thstandthe B + voltage applied to the tube. If a commercial

or surplus bulkhead capacitor cannot be found, thiscapacitor can be hame-made using a 1-1 2- inch138rnrn)square of sheet metal and sheet mica.

The plate tuning capacitor, C4, requires abou t 10

pF and m ust not break do wn at fu l l applied B + Alow loss, h igh quality glass or ceramic piston tr immer(rated accord inglyl wo uld do nicely, or this capacitormay be constructed using 314-inch 119rnrnl wide,sheet brass strap bent up near the plate l ine fo r about3 4 inch (19mm l and insulated from it with a layer ortwo of sheet mica see fig. 71. The spacing betweenthis tab and the plate l ine can be varied by any con-venient mecha nical means to tune the l ine. An alter-nate approach is to drill and tap the transverter top-plate to pass a n o . 8 or n o . 10 CM4-M5) brassmachine screw wi t h capt ive lock nut. Cut the screwso i t is just short of touch ing the p late line by at least1116 inch or 1.5rnm) and solder a 314 inch 119mmldiameter brass disc to i ts end. Insulate the disc f r o mthe plate l ine with mica sheet.Output power is inductively coupled from the plateline via L5 which is made f rom F/4- inch 6.5rnt-n)wide copper or brass strap soldered to the outputconnector IN or BNC type) and run parallel to theplate l ine about f 8 i nch l3mrnl away. The end ofthis strap can then be r un dow n to , and parallel with ,the chassis to form the matching capacitor C5. Alow-loss glass or ceramic piston tr immer of about 10pF capable of handling some moderately large r cur-rents, cou ld also be used.construction

The grid i - f and cathode LO circuitry plus f i lamentand grid bias wir ing are al l contained wi thin landshielded by a smal l aluminum minibox mounted ontop of the plate l ine enclosure. This minibox issecured by the same hardware which is used tomo un t the grid bypass plate, C7 Two screws are in-sulated wi th nylon bushings from this gr id bypassplate; the third screw makes contact with C7 but isst i l l insulated from the chassis by nylon bushings.Bias and i - f connections are then made to this screw.

The plate c i rcui try and output coupl ing loop are

flg. 6. PImm llns U in flgm. 1 and 2) for the I s M z tnnmvsrtur.nngsr .rock can be obtalnd from l m t r u m ~ t pselmftim. LlM.Falls. N-Jenny.


6/8

and air source for the power supplies and LO chaincomponents.Common vhf construction practice should be

followed throughout including short componentleads. quality component selection, and rigidmechanical assembly. In one case a 1116-inchE .Srnm) thick 4-inch IlOcrnE square brass plate was

An alternate awmngement for the output circuit.

contained within a 4 x 4 x 2 inch /10xlOx5cmminibox which serves as the chassis base. Screenedair holes provide a path for forced air cooling of thetube anode structure, which is absolutely essential atreasonable power levels. These holes must be rftight, so the shielding screen must be well groundedaround its periphery.

Copper screen can be soldered to the aluminum

Conatruttlon dmmlh 1 ?ha grid bypaus p b t a ahowing the sheetmica insulator cur out with an Execto knife and rhn Inaulnrlngwashem

Plate Ilne ramovd to show grld bypsss plat . fhlr ptnte llns ismounted on the two Teflon plllars t bottom

chassis y first tinning the aluminum with a largesaldering gun or iron. Generously coat the peripheryof the air hole with a heavy oil (clean auto engineoil works fine) to keep the aluminum from beingexposed to air, then sand the surface clean using or-dinary sandpaper or emery paper. Once cleaned, ahot iron 1200 atts minimum) and rosin-core solderwill tin the area beautifully. Then the copper screencan be soldered to the aluminum chassis. Practicethis procedure first on some smallaluminum scrap

A convenient way to provide cooling air is tomount the transverter on edge on a large, air-tightpressurized chassis. The air hole in the transverterplate-line enclosure should be placed over an equal-sized hole in the pressurized chassis.

This chassis can also serve as a mounting platform

used as the top plate of the plate line enclosure withgood results. It is a good idea to sand bare the adjoin-ing surfaces of the plate line enclosure, top plate,and bottom plate and use additional screws toassemble the top and bottom to assure good

PCATC W L 4 T C L I K PLATE L M L 4rrgnm?I

fig 7. fh rw methods lor bulldlng the plata tuning capaeltor U

shielding. When assembly is complete, a close visualinspection and a VOM continuity check should revealany obvious problems beforeapplying power.tuneup and adjustment

Initial testing can best be done using variable-


7/8

~ l a ~ l i n shomFng the plnta callst Ifingar stock) and eat mleainsulator, right.

voltage power supplies to provide the plate voltageand grid bias. Since rhe grid will typically draw morethan 50 mA, the bi s supply must h ve low im-pedance and be c p ble ofmainteining consr nt output voltage under fluctuating oad conditions. Applyfilament voltage and cooling air, and after adequatefilament warmup, gradually apply B + while wat-ching the plate current. Plate current should risegradually, indicating normal tube conduction. In-crease grid bias to reduce plate current until a platevoltage of about 300 volts can e applied with suffi-cient grid bias to limit plate current toabout20 mA.

fig. 8. HsFf-wavelength rsnsmlssion-tine fllter for use wlth the 1296MHz rrsnsvsrtsr.

Under these conditions, gradually apply the LOdrive to the cathode circuit. The presence of thissignal will immediately be indicated by increasedplate current. While not allowing plate current to riseabove 100 mA, tune C1, C3, and the LO chain formaximum plate current. Increase grid bias as re-quired to limit prate current to a safe value.A t this point, depending on your LO frequency.you mi ht beable to resonate the plate circuit to theL O frequency by adjusting C4. Thus the unitoperates as an amplifier to verify the behavior of theplate and output circuitry. If your LQ is above 1296MHz, tune toward minimum capacitance; if your LOis below 1296, tune toward maximum capacitance.

Adjust C4, and the position of L5 and C5 for max-imum indicated output power a t the L O fre-quency.Once cathode-line tuneup a t the LO frequency hasbeen accomplished and optimized, increase grid biasto reduce plate current to near cutoff about 10 to20 mA. Leave this grid bias at this value; the objec-tive of this procedure is to bias the tube a t or nearcutoff classA 8 or B) with plate voltageand LO driveappliedbut no i-f signal present. Now gradually applya carrier a t the i-f input and tune L2 for maximumplate current. Again, the presence of the i-f signalshould immediately be indicated by increased pratecurrent.If all's well a t this point, the LO and i-f signals arepresent in the tube s plate line: all that remains is totune the plate line to. the desired mixer product withC4. This process can be greatly simplified by placing

low loss filter tuned to 1296 MHz a t the output ofth transverter. An example filter of simple construc-tion is shown in fig. 8.6The filter can be pre-tuned to 296 MHz by placing

it in series with your 1296-MHzreceiving system andtuning for maximum received signal from a ne rby1296-MHz transmitter, the third-harmonic of a 432-MHz transmitter, the 4th harmonic from a 144-MHztransmitter, or any other convenient rf signal source.This assures that any transverter output obsewedwith the filter present will beon the desired signal fre-quency, and that power measurements will representtrue power at the desired signal frequency, not thesum of the power contained in all th trans-verter s mixer products

It is wise to leave the 'bandpass ilter in the systemto assure clean output. If, as is usually the case, yourLO is lower than the desired mixer product on 1296MHz simply back out C4, (tuning toward minimumcapacitance or higher frequency while watching theoutput of the transverter for a peak. Once this peak isfound, tune all screws for maximum output power. Itmay require a fair amount of time and patience to getthe feel far the effects of each adjustment, and mayeven require repeating the entire procedure with a

l h lute tuning empotzitor. C4 mounted under tha plnm Ilnetwwn the MOT dlon pillan.


8/8

Construction of the home-made plate rvnrng cupacttor. C4.

number of available tubes some fly and somedon'tOnce uneup under reduced power conditions hasbeen accomplished, full 0 -k m y be applied; valuesfrom 5 to 1000 volts have been used successfully,but with lots .of forced air cooling The tube mustthen be re-biased for an idling plate current of 10 to2 mA with LO but no i-f signal applied. Then in-crease CW i drive power up to the point of satura-tion (transverter output no longer increases with in-creasing amounts of i-f drive) and tune all adjust-ments for maximum output power. In the ssb mode,talk the transverter output up to an average of abouthalf the maximum CW power output. Typical stageoperating parameters are shown En table 2. The loweff cieney is Wpical for this type of mixer circuit.

The half-wave bandpass filter can also be used totransform a =-ohm transmitter output impedancefa75 ohms to match the impedance of inexpensive,tnblm 2. Typl-l operating paramwra for the 1296-MHztranuuerter.Output powar wan mwmur6d t tha output of the bandpama filterwith a callbrsted 81rd wenm&er and 8[IE slug.

L o ~ B I - o s c ~ ~ ~ B ~ M) MIBP 3wattscanier1 f powen 4 w a r n CWPlstavoltage +BWVdcPlate current F 5 0 mARt pwver input 120~ tG d i m ~ - 35 VdcGrid current SO mAR f m r zput 17waffsEtfrcienw 14%

low-loss C TV coax. For 75 ohms simply space thetap 5/8 inch (16mt-n)from the line end.conclusion

I t is the intent of this article to describe a relativelysimple and inexpensive method for the average

amateur to get on I296 MHz with enough power tomake himself heard. Hopefully the ease with whichthis system can be put together will encourage moreactivity on the band and finally put to rest the myththat you c_an'tget there from here on 1296 'V amwaiting now to hear from some hard-working anddedicated uhf buffs in Hawaii who would be in-terested in destroying the current 1296-MHz ter-restrial record once and for all How about it outthere, any takers?acknowledgementsSincerest thanks are in order to th many people

who contributed to this effort: notably to Bi l lJungwifih, WAGNRV, for the beautiful constructionof the first working prototype model of thetransverter; to Tom Staller, WBGQHF. for the con-tribution of a commercial varactor tripler; and to the

Hnfl-wavelength tran8rnl~#fon+lin~lft@rfor M MHz. Lot * o lZha llltsr shown here ia0.4dB. Olrnmnmlansere ahown In fig. 8 .

West Coast's Father of 1296, Bill Troetschel,KGWQH, who provided much good advice and en-couragement amongst many good ribbings Photocredits go to Alan Monie. Special thanks go to mywife, Jean, for typing the manuscript.

references1. Ertward Tilton, W1HDQ, Building Lo w Cw t R Attanuatm.QST, May. 196B. page20.2. Wll iam Trcwtschel, KGUQH. 1296 Rwisited, QST. July.1973, pagem.3, Jmeph Moraski, K4StJM. and Charles Spie W4API, A Fre-quency Tripler for 1296 MHz, rn radio Septemhr, 1969,page404. Frank Jonas. WGAJF. VHrlor the edioAmeteLlr Cowen Pub-lishing, Pon Washington, New York. 1958.5. 0 . Evans. GSAPE, and G. R . Jessop, G a P , VHF UHFManuel Radio Society of Grear Britain, London, 1976, chapter 5.6 . Paul Magee W3AED. Quad YagE Arrays for 432 and 1296MHz, hem radio M a y , 1973. rl filter on pagea ham radio

Documents

1296 MHz Transverter