Web: http://www.pearl-hifi.com 86008, 2106 33 Ave. SW, Calgary, AB; CAN T2T 1Z6 E-mail: [email protected] Ph: +.1.403.244.4434 Fx: +.1.403.245.4456 Inc. Perkins Electro-Acoustic Research Lab, Inc. ❦ Engineering and Intuition Serving the Soul of Music
Please note that the links in the PEARL logotype above are “live” and can be used to direct your web browser to our site or to open an e-mail message window addressed to ourselves. To view our item listings on eBay, click here. To see the feedback we have left for our customers, click here.
This document has been prepared as a public service . Any and all trademarks and logotypes used herein are the property of their owners. It is our intent to provide this document in accordance with the stipulations with respect to “fair use” as delineated in Copyrights - Chapter 1: Subject Matter and Scope of Copyright; Sec. 107. Limitations on exclusive rights: Fair Use. Public access to copy of this document is provided on the website of Cornell Law School at http://www4.law.cornell.edu/uscode/17/107.html and is here reproduced below:
Sec. 107. - Limitations on exclusive rights: Fair Use
Notwithstanding the provisions of sections 106 and 106A, the fair use of a copyrighted work, includ- ing such use by reproduction in copies or phono records or by any other means specified by that section, for purposes such as criticism, comment, news reporting, teaching (including multiple copies for class- room use), scholarship, or research, is not an infringement of copyright. In determining whether the use made of a work in any particular case is a fair use the factors to be considered shall include:
1 - the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes;
2 - the nature of the copyrighted work;
3 - the amount and substantiality of the portion used in relation to the copy- righted work as a whole; and
4 - the effect of the use upon the potential market for or value of the copy- righted work.
The fact that a work is unpublished shall not itself bar a finding of fair use if such finding is made upon consideration of all the above factors
♦ PDF Cover Page ♦ ♦ Verso Filler Page ♦ MODERNIZE '10"" 'W� II�
HE of extreme linearity. hs prefl:rrN oper· T Williamson amplifier circuit was performance is required. Even this mini· hrst publicized in Enlland in 1947, mum will probably be: inctcaScd in the ating conJition is as a pentO\lc: ;although .10.1 in Ihis counny in 1949. It has ycan to come unless the c:fficiency of the: manufacturer furnishes trio.. le r;acin�s .u:hievc. .. 1 wiJe acceptance and popularity, loudspeaker sYStems can he increascd. for che tube. rri,,..le UI'l'r;alion r�'sults in .111.1 lI.n 1",,,11 the basis for K'veral modi· Por theM: Itnd relatN reasons. c:lfuru higher .JislOrriulI an,1 rl",h• .:e.. 1 powl"r vut· hl'ations of the: ori,inal desian. The: have bcc:n devoted to increasinl, the: put. The convencional ultra· linear con
. · mnS( billsic change was the: ulrra-lincar power OUtput of audio amplifien. The ncecion cannot provide opcimum rc:sults version of operation, which I developed advent of some new tube: t}·pes ha', made wich dICse tubes either, since there is no anJ subsequc:ntly Jc:scribcd• . This ar this praCtical within the W:lliamson type of operation more linear than che rangement corrected 2 of the: basic de configuration without the n�cd for com· pentode conneccion for which chey were he ic:ncic:s in the: original desiln - it pletcly tebuilding the amplifier. Changes designed. As will be: discussed later, increased the: powc:r caS-bility of the: required arc replacemc:ne of the output however. a compromise form of opera· amplifier to n or 30 wattS, and it irn tubes, substitution of an output trans· tion fits the necJs of che Williamson proVN the: marlin of feedback stability. former which will handle the: increased modernization very nicely. Now, as always happens. Pro,I'CSS in power and provide suitable impedance To dc:rive the: potential be:nefits avail· amplifier Jesi,n has continued: it is matchin& and addition of fixed bias. able in these: tubcs, proper impedance pussible to make furthe:r improvements macching muse be: obtainc."! in Ihe out· in 'the Williamson Jcsigrf'(both triode , New Output Tubes put transformer, The Ornaco A-430 and ultra·linear venions). These im The: new tube: selected for modernization transformer has bc:c:n dc:signed specifi· povements again correa for limitations of the: Williamson is che Amperex cally for chis purpose. This is a �O-watt with respect to power OUtput and ('CA7, which is also imponed anJ dis unit. che performance of which excI:cds stability. tributed as che: Mullard EL· H. This is Mr. WiIliamson's sIX-cificarions wich re· a compact tube: with power capabilities spect to frequency response. permissible Increasing Power Output up to 100 wans, depending on the sup· feedback. power handling ability. ;and so Present thinking 00 requirements for ply voltagcs available. Ir can be: plulled on. At presenc this is the only commc:r· .IuJio P�W" js vastly dilfercnt from that ditcCdy into che sockm formerly uscd cial transformer of corrt'Ct impl-Jance. nf a few yean ago. Then, most people for �RRI',. KT(,(,'s, 161<'1·s. ;an.! mher hut it is anticipate.1 thOlt orhl"rs may he: , .. id, · .... l"n waus is enoush for me." tubes of this type, with the single: aJJi· available soon. Now, however, modern program ma tional requitement that the No. I pins The Dynaco A·4.30 has primary taps terial has bcc:n increased in dynamic muse be: grounded. which can be: used to furnish aboue 10% range many times over that of former The: 6CA7 JEL. 34 is a pentode tube: screen loading. This does noe cause de· �l:ars. This fact alone: has increased the: power requirements substantially for realistic, undistoncd teproduCtion. In addition, source material frequency re sponse has been cxtended, and this also introduces the: need for a te-cvaluation lIf amplifier power requirements. In crcascJ frequency response means that the amplifier has to handle power at greater ntrcmes of frequcncy. At these: e xtremes, the impedance characteristics of the loudspeaker change from the: nominal values. This means that the amplific:r is mismlllcheJ at frequency ex· tremes, and a mismatch dccrc:ascs the maximum.powc:r cap.bilities of any ;amplific:r. To Jc:liver clean power into a loud 'Ilt·,.lel·r IlIa di AUIlIOCltAFT I\fACAZINE hi8hct ratil1�) arc \V1f�-J III serieS i fom this junction, with the negative rl"Ctilier I 'l'rminal� tow;lr,1 rhl." ""trut �i,'(·. Th .. rl·s:.rhinH negative OC is filtl."rl-J by a IO.OOO-ohm potcmiomercr and a 47,· OOO-ohm fixed resistor, and the arm of the pot is bypassed by • 40-p.fd (or greater ) , I �O·volt Cllpacitor. This is shown in the diagram. Fig. 1. The po tc:ntiomerer can be placl-J conveniently in the hole that formerly hcld the bias· balancing pot. A bias-balance adjust. ment is no lonlter required, since the OUtput transformer is of • design in which performance is noc deteriorat�-d by moderate current unbalance, Ind the rubes used do not have much vari;!tio:l in plate current drain. The new potentiometer controls the bias voltage, which is fed to the bottom ends of the two IOO,OOO·ohm outpUt. tube grid resistors. These, of course, are no longer connected to the components formerly used in bias balancing. The combination of chan,ltes descriheti above has inc:rC';lSc:d the !'uwer ut the i amplifier to about twice iu ultra· lincar ·i rating and about 4 times iu trioJe rating. This dunge alone makes In im· po"ant improvement, but an equally important improvement can also be , made by extending the stability margin i of the amplifier. J Stabilizing the Amplifier Criteria for good amplifier design have changed in recent years, and the stress is now being placed more and more on :rioration of the extreme line:uity of plifier) . It fits the space, despite its lar.fl:e am plifier stability. Many amplifiers, t s, and it does h ve the advantage size, because it hu no flanges. Insr.alla· the ube a while performing well under steady-state of lowering the nternal i ance tio the transformer requires no me i mped (for n of conditions, have exhibited muddy and damping). further, it improves chanical alterations in the chassis except, . bener harsh qualities when reproducing music. the inherent regul.uion of the output possibly, reaming the mo nt ng holes to u i One reason for this is the fact th;!t stage to he point where no changes need acco modate mounting sruds. t m their transient performance is inferior made in the B+ supply of the basic be to that under steady·state conditions. Williamson in order to the new Biasing the Output Tubes use A.nother reason is that amplifier per· tubes. The 6CA7 34 requires a lower value JEL- formance on loudspeaker loads is often This transformer can he interchan,lted of Ilias than rubes generally used in not as good as it is with resistor loa.J� Williamson amplifiers. Value of the . .Iir��rly with ullits fnrmerly u5<.·,1 in this This point was touched on briefly he· c r it. If the uri}o;in;11 output ruhes were catho.le resistor should he reduced to i cu fore: its ramifications with respect to ahout 200 ohms from the conventional triade connected, the 100·ohm screen feedback instabiliry are far· reaching. suppressors connected from pin 3 to pin 2�0 to 300 ohms. When this hu been Many designers have come to the con trans done, after substit ting rubes and out 4 should be removed. and the u elusion that stability, has more effect on p t transformer, the resulting amplifier former leads connectt:d as indicated in u listening qU:llity than disto"ion does' e can put out 3� to 40 exceptionally clean th schematic ,!iagram (Fig. I). The ConK-quently. increases in the m:lr,ICin of constructor should note that the circuit warts. However, the capabilities of the must be traced from the: phase inverter new rubes and transformer are not fully to the output grids in order to ,Ietermine exploited unless the constructor is will· which of the output tu�S is the "top" ing to incorporate a fixc 17 , �bilit�· arc important Jcsi,!;n prublc:ms. raplJly IOwarJ il yanishi,,� point. The: t:�UlppCJ; jf n(ll, onc: .houlJ i>c .lJ,kJ. Br rhe51: sranJar,ls, rhen, original frequency rl'Sponse of the amplifier will The .05-p-fd couplin,!; capacitors bc: WiII,;m�n amplifiers hav� ina.lequar� be approximately the same as that of the tWttn the two 6SN7's should be in· . srabilir�' ;lr b oth l'xrreme IlIw anJ hi/-th original version el[cept that peaks in the crl'ascd to O.2� p-f,1. Car� sholll,1 he frnlll"1Il i .. �. Thi, UII 1><: ,1C:1ll01l5Ira"... 1 rc:spUllllC (a",)(iaft.� with instllhility) cxcn:isc.-..l in ord�·c to get thc:lIC 10 hI fnr the.' IlIw ('11,1 hy IlIuchin" the: inllUr a� eliminated. The rransient responle the SI'l\lCt'. If di/liculry is enCllunrecl.'d. ic grid momentarily with the: finge"ip and - thar unmeasurable intangible - will is slJ.&:,I:esred that miniatu� capacitors be watching the spt.-:aker con�. The heavy be audibly better. It will be particularly used. such as Aemlitc.'S made by Aerovox. low-frequency transient which is gen �vident in mo� solid, better·defined The l·megohm phasc-corCl't:tins rc:· crated tri.ltgcrs the ilml'lilier, and ther� bass and smoother, cleaner rreble. sistors across the coupli_ng capa.citocs pt.·c· ar� usually sevcral sur�c:s befor�/ th� mit som� positive DC to appear on rhe Appendix dfl'Ct is Jamrc:J out. The: speaker' con� grids if the negative bias supply is in· can be ob5l:rved to mov� back ar.d fo"h The following s pecific hints will be orc:rative. The existence of nt·,r.::llive 5I:veral timl's hdllre comin,IC t ... rc:st. This hc:lpful ro those who arc intereste.-d in voltage at the: grid should be chl'l.kc:·,1 means that �hnrt siltnal impulses will modernizing the Heathkit W-3M Wit with the rccrifier removed bc:fore �r· also cause: sl'urious cone movcmcnrs liamson: mitting the B+ voltage to he a pplied. which tend to hlur the sound. The bias voltage divider an.d rectifier Then the rectifi�r should be inse"l·d and At high fn:quencies, the correspond can go in the power supply chassis. Out the bias set to 3' volts from �rid to ing effect can he viewClI on an oscillo put of th� selenium rectifier is connected ground ../In the cubes have had timc to scope with a square-wave signal input. to pin 5 of the pow�r socket and carried warm up. Do not remov� the secon,1 A ripp lc...J square wave is indicative of rh rough the spare wi� in the connecting 6SN7 while the amplifier is on, because hasic instahility, and in,licates rhat cable to one side of the 10,OOO-ohm this will cause additional positive volla,r.:e rhere is a transient distllrrion of high bias·setting pot. to be applied to the grids of the out frequency signals. The 2�O ·ohm and twO I OO·ohm resis put stage, upsetting the bias and possihly Insrabilitr is due to rhe facc rhac tors are discardc:d, as is the lOO-ohm pot. harming the cubes. the phase: characteristics of th� amplifier The bias pot �places the lOO-ohm bal C:1U� some IIf rhe fl'Cllt-ack ro he al'l'li, ... 1 ance :l,fjustment, and the hlank tie AUDIOCRAFT Test Results pnrilil·..J, illsrc:hl of Ht'X.,I;,·d, at the: puints to which the WO-ohm resistors The amplifier shown in Fip. 2 and I, fn:quenc)' exrrc:mes. The n:me.�y is . "'e� fastened can then be usc...J ro con which was the basic Williamson con · superficially simple - to shift the: phase nl'Ct lead. which a� pan of the ground verted in accordance with this a"icl�·. in the right direction at the critical f� circuit. produced 0.6% IM disto"ion at �o quenci�s. le is not always simple to do In the intercsu of econom)" the watts output. 0.2% at 36 watrs, and rhis. Forrunarcl�·. the phase characteris 20'p-CJ capacitor which formerly by- 0.1 % at U watts. Below I � watts dis tics of the A-430 transformer and the to"ion was in the residual range of the \,(/iIliamson circuit arrangement permit meter and could reasonably be called complete corn:ction of the low-frequ�ncy negligible if not unmeasurable. Test -p1r.UeChar,u:tcristic',and apr>reciabl� -cor frequencies we� 60 and 7,000 Cp5. rection of the high-frequency characc�r mixed in a 4-ro-l ratio. iscic. These corrections a� made with a These: readings could have been im· few inexpensiv� componenrs. proved slightly by using pans matched The low-frequency correction is mo� precise:ly, or by adding to the achieved by shunring the 0.2�-,.tfd power supply filtering. It is doubtful s a improvement coupling capacitors which go co the out that such m ll would he passed the cathodes can be used for put g rids with I megohm resistors_ apparent audibly. On the orh�r hand. if filtering the bias supply. The schematic the phase inve"er balance w�re off, Ot High-frequency correction is obtained calls for 40 p-fd, but the difference in certain other �Iements had drifted in with a 100-p-p-fd capacitor which is COD hum level is only 2 db. value the disto"ion figures might be nected from the lower driver plat� to , Jacks formerly used for metering plate doubled - still exceptionally good per· rhe cathode of the firsr stage. Without current can still be used to check equal formance. ,ltoing inro the theory underlying these: ity nf the output tubes. The former con <.t'rr,·cti,,"!, il is wurth IIICllliollill1( Ih:lt frequency rnpnn� was perfl'Criy 1Ii11 ",:ction to the 2'0-ohm bias resistor within the range of our test equipment. they have a trem�ndous effect on per must now be: grounded in ord�r co com Calculated response, without the 100- formance. ( It is assumed that the plete the circuit path for the cath-xle cur p-p-fd f�baclr: capacitor at the driver 10,OOO-ohm resistor in the input grid, rent. 'At the same time the No. 1 pins stage plat�, is db from 2 to 200.000 change of the .0�-p-fJ capacitors to 0.25, ± I ( suppressor grids of the output cubes) cps; with this added, response: above and the use: of. a small capacitor across should be connected at the socket to 80,000 cps slopes off smoothly. This the feedback resistor as indicated in the the No. 8 (Cilthode ) pins. capacitor causes a very slight increase: in schematic are already included in the In order to inse" the 10,OOO-ohm distortion at 20,000 cps which is quite amplifier. If not, these should also be , parasitic suppressor resistor directly at i signifi nt compared to th� 12 db in· added in accor,lanc� with previous n ca the input grid, the 2 2 megohm resistor a in ta t recommendations·. ) . - cre se: the high-frequ�ncy s bili y should be reconnected directly from the margin. Stability at the low end, deter· This completcs the modernization of input socket to ground. The 10.000- mined by recovery characteristics f corn the Williamson. If the power supply purs uhm n:sistor can then be inserted from out a full 150 volts with �nable regu a sharp overload pluse, was. apparentl}' the input connector to pin 1 of the perfect. lation, the output pow�r will be about first ('SN7 tube. This replaccs the .05- Squa�-wave response: at low frequcn. 50 wattS at 1% IM distortion. If the p.fJ capacitor which is used as part of cies was excellent, and �xceptionalh' power supply provides lower voltage, the Ihe bias supply iD the power chassis. good at high fCl'quencil.'s. Power re· output power will be rcducl-d somewhat. Make sure that the pre:amplifier used has sponse: was completely flat from 20 III Below full output, the distonion Jrops an output coupling capacitor since the 20,000 cps at '0 watts. Toral COSt for amplifier now has none in the input. all parts required to ma ke Ihe complt·te °H.ner. D .• "A 60 Wen Uhre·Li""... AmpIiIia", H",/i. .. r.I ...-;., .. N._. FebnIM,. IIISS. Practically all preamplifil.'cs are so conversion: less than $·10.00. AlIl>IO(RAI'T MAGAZINE