sign in sign up
<<
Home , Vacuum tube

A PRACTICAL APPROACH TO TRANSIST TUBE

BY F. J. BECKETT , INC. ELECTRONIC INSTRUMENTATION GROUP DISPLAY DEVICES DEVELOPMENT

Two articles published in past issues of Service Scope contained information that, in our experience, is of particular benefit in analyzing circuits. The first article was "Sim- plifying Linear- Analysis" (issue #29, December, 1964). It describes a method for doing an adequate circuit analysis for trouble-shooting or evaluation pur- poses on transistor circuits. It employs "Transresistance" concept rather than the compli- cated characteristic-family parameters. The second article was "Understanding and Using Th6venin's Theorem" (issue #40, October, 1966). It offers a step-by-step explanation on how to apply the principles of ThBvenin's Theorem to analyze and understand how a circuit operates. Now we present three articles that will offer a practical approach to transistor and vacuum-tube amplifiers based on a simple DC analysis. These articles will, by virtue of additional information and tying together of some loose ends, combine and bring into better focus the concepts of "Transresistance." Part I THE TRANSISTOR AMPLIFIER INTRODUCTION

Tubes and are often used to- there is an analogy between the two. It 50, it seems pointless not to pursue this ap- gether to achieve a particular result. Vacu- is true of course, that the two are entire- proach to its logical conclusion. um tubes still serve an important role in ly different in concept; but, so often we In this first article we will look at a and will do so for many years come across a situation where one can be transistor amplifier as a simple DC model ; to come despite a determined move towards explained in terms of the other that it is and then, in the second article, look at a solid state circuits. very desirable to recognize this fact vacuum-tube amplifier in a similar light. We will assume that both devices are op- Whether a circuit is designed around Transistor and data give us erated as linear amplifiers and then use vacuum tubes or transistors or both, it is very little help in the practical sense. Pa- the results in a practical way. important to recognize the fact that the rameter Curves and electrical data show One must bear in mind that this ap- two are in many ways complementary. It the behavior of these devices under very proach cannot be assumed in all cases. is wrong to divorce vacuum tubes and defined conditions. In short, they are more It is, as it is meant to be, a simple analy- transistors as separate identities each pe- useful to the designer than the technician. sis but the results will prove to be a valu- culiar to their own mode of operation. \Ye are often reduced to explaining most able tool in trouble-shooting and under- Indeed, as this series of articles will sho~ circuits in terms of an ohms law approach; standing circuits.

Let us consider the general equation fo~ One is quite justified in looking at a current through a P.N. junction. ORJECTIVE transistor in terms of the two-diode con- cept, refer to Figure 3. Therefore, assum- v The objective of this paper is to present (1) ing diode A to be forward biased and di- a practical approach to Transistor and ode B to be reverse biased, as would be the Vacuum-tube amplifiers based on a simple where V = applied volts case if we were to operate the transistor DC analysis. I, = reverse bias current as a , the current through = constant between 1 & 2 The articles will be published in the fol- diode A will conform to equation (1). kT lowing sequence. Let us take a closer look at Figure 2. and V, = - where k = Boltzmans cl 1. The Transistor Amplifier. We define conductance in the gener*, Const., 1.38 s lO~'~oule/"Kelvin 2. The Vacuum-tube Amplifier. case as I T = absolute temperature in degree Kel- 3. An analysis of a typical Tektronix hy- g=- vin at room temperature, i.e., T = 300°K brid circuit (Type 545B vertical) based v on conclusions reached in (1) and (2). q = electronic charge 1.602 x 10-'You- and therefore at our operating point "A" the dynamic conductance lo111b. As a corollary they will bring forward 300 some important relationships between vacu- V, r -= 0.026 volts um tubes and transistors. 11600 hence v I, esp --- pV, g' = pVc

Figure 1

Figure 1 shows a typical forward volt/ amp characteristic for germanium and sili- con . Figure 2 is a plot of the col- lector current or the base cyrrent versus Figure 2. Line (1 ) is a plot of the base cur- the base-to-emitter of a transistor; rent versus the base-to-emitter voltage (VBE). Line (2) is o plot of the collector current point A on this curve is a typical operating versus the base-to-emitter voltage (VBE). Point Figure 3. Illustration of the two-diode con- point. "A" is a typical operating point. cept of o transistor.

@ 1967, Tektronix, Inc. All Rights Reserved. I tlie antl the common I~aw sisted of resistances alone, it would be pas- but I 10 tlien g' = - >> ,,,V,> coniigurations. sive; i.e., it could supply no energy of its own. But a transistor can amplify energy Firstly, let us define tlie tern] p (tlie sm:tll-signal current gain) as to the signal. To represent tliis we liave sl~o\vn a current generator shunting R,. AI<. The term takes into account the re- p = -- The value of R, will depend 011 tlie circuit coml)ination of carriers in the junction re- AT,, configuration; i.e., tens of kiloh~iis for a gion It is approximately unity for ger- :tnd since In = I, -t 11, comnion emitter configuration, to many tnaniuni antl ;~pprosiniately 2 for megolinis for a common base configuration. At a typical operating point tliis term can 111 our approach it is not necessary to pur- mually be neglected. Therefore, we may sue this matter any further since we will say that usually p >> 1 tlien 11: =: 1, not he considering a transistor in any ex- treme condition.

Now resistance is tlie reciprocal of con- Kow in a more pr:tctical sense, let us tlie enlitter current flows into tlie hase. ductance antl therefore the value of con- look at Figure 5, a typical common-emitter Hence, it is reasonable to suppose that any (luctance at point "A" can be given In term\ con£iguration. impedance in tlie emitter, wlien viewed from of resistance tlie lnse, will be p times as great; and, :my impetlance in tlie base, wlien viewed iron1 tlie emitter, will be p times as small. This resistance (r,) is commonly laio\vn Tliat is to say, the dynamic resistance mul- as the dynamic emitter resistance. tiplied 1)y ,Cj niust equal R, in our equiva- lent "Tee" circuit. At this point we will tlepart from our simple model and look at the transistor in Hence R,. = pre another form; but, Ixar in mind our first Our equivalclit circuit slio\vs :t resistance thoughts. Transistor parameters are derived It,,. This resistance is known as the base- from various equivalent circuits depending sprcatling resistance. It is :L physical qun~i- upon tlie configuration i.e., common emitter, tity and can Ix expressed in terms of resis- comnion base, or . We will tivity associated with tlie Ixw-emitter junc- not consider any detailed analysis in this lion. It can vary hetween :t few olinis to approach ; hut, to untlerstantl tlie approach Iiuntlretls of ohms, depending upon the type :f is necessary to know how tliese param- of transistor ; :tnd therefore, nus st be taken 2rs are derived. It will 11e simple enough Figure 5. A typical common-emitter circuit. into consit1er:ttion. T.ooking into tlie emitter to derive another set of parameters once we see it ns an impeclance wliose value is we liave our basic model constructed. tlivitled Ily P :tnd appears in series wit11 The simplest and easiest equivalent cir- the t1yn:lmic emitter resistance (r,.) . Hence cuit of a transistor is the "Tee" equivalent. the emitter current encounters an impetlancc Tt is a very good approximation about the in tlie I):lse/emittcr junction wliicli is equal l~eliavior of a transistor, especially at DC lo tlie sum of tlie dynaniic resistance plus and low frequencies. \\re can also represent RI, either tlie comnion emitter or the common - , the latter term we will designate R, lnse simply hy interchanging RI, and Re. P ,Ihe. input impetlance we see looking into Figure 1 is a "Tee" equivalent circuit of and the sum of tliese two resistances we the base of a transistor in tlie conimon emit- will designate Rt. ter co~ifiguration is

The value of R, can vary anywhere be- EMITTER Re Rc COLLECTOR tween 2 R to 24 R depending on tlie value AV,,,, also 411, = --- of RI,. RI, is difficult to measure and rare- RI,, ly given in electrical data 0x1 transistors. A figure of 250 R's is a typical value at low BASE frequencies. Therefore, if P were 50 tlien R, \vould he 5 R's. COMMON BASE we also recall that Now if we look into tlie base in tlie com- mon emitter or the comnion collector con- p = aIc i~guration it is reasonable to suppose we art, will see tlie resistance (Rt)-plus any other impedance which may be wired to the emit- ter terminal-multiplied hy P, then Tlierefore substituting equation (13) in RI,, = P(Rt -1 RI:) (10) equation (14)

EMITTER where R,: = the external emitter resistance. If Rli >> Rt then Ri,, = PRr: COMMON EMITTER So far \ve have had very little to say ;~ndfrom equation (15) Figure 4. "Tee" equivalent circuits for the allout R, shunted by tlie current generator common-bose and common-emitter configuro- tions of transistors. I,:. If our equivalent "Tee" circuit con- av,,, = ~1~ (RI: + RI) (16) we define the volt;tge gain as can be placed in this category so far :is the signal is concerned.

The apparent AC ground may be rep, Then from eclnation (12) and equation sented by any point in a circuit \\-hich acts (16) as to represent a low impedance betwen that point and the actual .4C ground thereby bypassing the signal to an actual rZC ground. A large value is a typical example should one side be returned to an actual AC ground.

The virtual AC ground point is perhaps the most difficult to recognize. It may best be explained as that point in a circuit where we have two signals of equal amplitude and frequency but exactly opposite in phase. If we analyze the comn~on-baseconfigura- Figure 6 will help clarify these points. tion in a si~nilarmanner we arrive at the Figure 8 sunimarizes the results of our same result with the one exception that the Figure 6. Illustrating the three types of AC DC analysis of the common emitter, com- sign is positive. ground normally encountered in electronic cir- cuits. mon base and common collector. The conclusion we can draw from this :tnalysis is that the gain of :I transistor stage is set by external conditions provitled that the emitter resistance is sufficietitly great enough to "swamp" our internal resist;tnce (12,). In the absence of an emitter resist- :mce

There is one very important 5:tct lie must renienilxr about R1:. RF will I)e that impedance In which the signal current will flow to the AC ground. \\'e define an AC ground point as that point in a circu~t at which the power level of the signal has been reduced to zero

\Ve normally encounter three types of an Figure 7. We define the porometer Rc in the common-base "Tee" configurotion as; ohms .4C ground : R, = -- A I< Where AVce is the change in the collector voltage because of the change in collector current Al,, when we hold the emitter current IEconstant. Once the collector becomes soturoted, the change in I, is very small for a large change in V,,. This is the chassis point or the I)C Hence, R, is o very large resistonce and does not modify the DC equivalent circuit to any extent. ground point. It is as well to remember the For this reason it was omitted from Figure 8. Therefore; Rout = RI, (Common Base). BASIC CIRCUIT EQUATIONS REMARKS

BASE COLLECTOR

A resistance IRE) between the emitter terminal of the transistor to the AC ground will modify the gain equation and the input im- pedance; then, Al?) = - -RL R, + RE R,, = /3 IRE + Ril. EMITTER

COMMON EMITTER

The equivalent resistance RE(,) between the input signal rource and the emitter terminal of the transistor will modify the gain fMlTTER COLLECTOR equation and the input imped- ance as seen from the signal source; then,

BASE

COMMON BASE

BASE The actual value of R,. will de- ROUT= pend on what resistance is con- nected to the bare. Let us ossume the base is directly coupled to the parallel with RE preceding stoge. The equivalent output impedance of the preced- ing stage becomes the numerator over beta in the second term in the parenthesis and the output impedance of the stoge under consideration Ltis modified oc- cordingly; e.g., if the output im- pedance of the previous stage is I000 r, then 1000 , Lt= IR, + -----I In parallel COLLECTOR P with RE.

COMMON OLLECTOR

Figure 8. PART 2 THE VACUUM TUBE AMPLIFIER In the previous article (Part I, "The Tratisistor Atnplifier) of tliis series, it was slio\vn tliat the gain of a linear tmn- sistor amplifier is set by external condi- tions. Tlie same reasoning can also be applied to \racuurii tubes. The equivalent circuit of a vacuu~ii-tubeamplifier is sIio\vn in Figure 9. Tlie current tliat is protlucctl in the plate circuit 11y the signal (E,) act- ing on the grid is taken into account by postulating tliat tlie plate circuit can be replaced by a generator, -pE, having an internal resistance (rp). \\'e may also consider a vacuutii-tube amplifier in terms of tlie constant-current form by replacing the voltage generator in the constmt-volt- age form \vith n current generator (gm E,) shunting tlie internal resistance (rp) . Figure 9. Illustrating the more familiar equivalent circuit of a vocuum tube amplifier. (a) The constant voltage generator form or the Thhvenin equivalent. These t\vo approaches are valid in every (b) The constant current generator form or the Norton equivalent. respect but they do not convey tnucli to us in the practical sense. Let us now con- 1 I (Grou+tded ) in terms of tlie date current. Therefore. sider a wcuum-tube ampliiier from an- to derive tlic actual gain figure we must other approach. We will now loolc :it a amplifier in terms rel;ttetl to our equivalent circuit. tleterminc tlie actual amount of cathode In an amplifier \vliicli has its grid ref- The common component is of course, the currcnt wliicli \vill finally reach the plate erenced to ground ;dl plate-circuit impctl- plate current. Tlie change in tliis current :ind 1)ecotiie sigd current. This figure ances, RI, and rp, when viewed from due to the action of a nil1 can be arrived at from n graphical analy- the cathode are multiplied 1)y the term determiiie the output voltage across the sis of the mutual-conducta~ice curves. In load impedance (RI,) . most cases, about 72% of tlie cathode cur- 1 rent reaches the plate to become signal . Also, by the same reasoning, tlie Now E, = E, B PS.1 + current. A typical example is a type 12BY7 That is to say . However, tliis figure can be as cathode impedatices when vicned from tlie high as 90% for some types-for exaniple plate circuit are multiplied by tlie term :I 7788 pentode. Tlie ratio of the plate ci. (p + 1). Therefore, tlie impedance we rent (I,,) to tlie cathode current (Ik) is i. see looliing into the catliode nus st be rp + Rl. + R, 11, or, E, = I, [(-) ] (20) plate efficiency factor, i.e., 7 c- -. --rp 'RL , where p equals the amplifica- I !, lL+l Now let is reexamine what effect this tion factor of the tube. fact must have on tlie gain of a pentode :unplifier :IS co~iiparcd to a triode ampli- Hence it is reasonable to suppose that and El, = -Il,RI, (23) fier. Tlie impetlnnce \ve see looking into the voltage E,, reference Figure 10, appe;tr? \Ve define tlie voltage gain A(,, as the catliotle of a pentode is the same as for across this impedance we see looking into a triode. the cathode. rp + 131, That is -- P+1 Iiowever rp > > liL and tlierefore RL can usu:illy be neglected in this equation. r 11 1 That is to say -- - -- ~+1- gm and since conductance is the reciprocal of re%istance we will call tliis impedance rr. \\re now have arrived at an equation for gain which is a ratio of impctlances. The same approacli may 11e applied to tlie grou~itlcd-grid configuration :uid \ve arrive \\'e have sccn tliat the gain equation of at a sitiiilar result, except the sign is posi- the triotle atnplifier is def~ncdin terms of tive. tlie parameters p and rp. \\'e should not lose sight oi tlie fact tliat p :md rp are related to the pl;ite current :ind tlierefore In tlic triotle amplifier a11 tlie catliotle \vhcn these pnr:i~neters :Ire tr:uisferred to current xvill ilow trhrougli tile output load cathode dimensions these terms must ' impctlance (RI,) . I-Io\vever, in the c:w of multiplied by the plate eiiicicncy f~tor( Figure 10. A vacuum tube amplifier in the ground the penlode and other multigrid tul~es,sonie That is to say the itnpetlnnce \vc see look- :d cathode configuration showing the various volt of tllis currcnt is diverted into tlie screen. ing into tllc c:~tliode r, must I)e multiplied 1ge measurements around the circuit. Equation (23) defines the output voltage 1)). (7). \\'it11 tliese facts in miiitl let us now derive the gain equation for n pentotle \\'here rh = ------+ Rr' (either v(,)or V(:)) ;~mplifier. P+l \Are recall that:

'= Ebb - El, --EL (22) \\'it11 a push-pull pentode amplifier we and E,, = -I,R!. (23) nlust consitlcr the plate-ef f icicncy factor ('I) Therefore, also E, = E, + Ek (19) . = ?~rhIhf IhRh (Z7) (34) but Ili = -111 (38) 'I Therefore substituting equation (28) in equation (27) where rk = --yeither VC1)or Val 'Irhr,, I]>Rh gm 1.3, = - i-7 17 Rk = 11, (1.k + ----) (29) n = plate-efficiency factor of ei- 'I Lllel- V(,) or V(3. and since the voltage gain Figure 11. A typical push-pull triode amplifier. We normally encounter two virtual AC ground points between the cothodes V, ond V?. It moy The c;iscode amplifier fr~~~tlanienl:~llycon- be necessary to consider the effect of the virtual sists of two tulm co~mectcd in series, see AC ground point at the iunction of R, and R?. If RI or R? is large in value compared respectively Figure 12. Sor~n:~lly\ye usu:llly fix the to Rk(1) or Rq?) then we con neglect this virtual grid of V(,)at sonic positive voltage. AC ground and consider Rli in terms of Rkcl) or Rlicl.). However, if this is not so, Rli will be the The key to ~intlersta~itli~~gthis type ol cir- Rr. porollel combination of RB(I) and R, or Rli(?) and R?. - - - (30) rh + Rh cuit is to co~lsiderV(?) as a voltage-activated current generator. A11 the current delivered I)p V(?) passes through the output load im- The same rem;lrlis \ve mndc :tl)out the es- pet1:iilce RI.. Any change in voltage xppear- terual enlitter Ri: (refer to I'art ing ;it the grid of V(?)appe:trs as :I ch:unge No. 1, The Tr:unsistor Ainpliiier) :~pply in current across ICI.. \\'e can derive the equ;illy xs \vcll to the cathotle resistor, Ra; gain eqr~ation in the s:me way as we did j~rly,Rk will 11r fliai i111pcd~711ccin xAir11 for a pentode amplifier. There is no need .< sigi~c~lcirrt-rnf ail1 flor~~fo file AC to consider ('I) if both tubes are . g~ozi~~d. 111 the case of the grounded plate (the cathotlc follo\\cr) we do not tleetl to con- 5ider the plate eff~cicncyfactor if the am- rpe3 where I-~(z)= -- plifier is triode connected, therefore, the P(?) + 1 "gain" can 1)e considered in terms of a simple divider network \vhich c:~never --- Ilc greater th;m unity. grnc.) where the sul)scripts (1) and (2) are as- sociated with V(,) and VcJ). One of the ad\.:mtnges of this type of Figure 12. Illustrating a amplifier usin circuit is t11;it the inter~~:dinipetlatice \vhicli wo triodes. shunts 111. is cxtretnely high.

In this respect the triode cascode ampli- +Ebb fier closely :~pproxi~~iatesa ~)entotlc ampli- fier. If we compare the plate-cr~rrcntversus pl:~te-voltage curves of 110th devices \vc see x close resemblance.

Figure 13 is a typical coniigulxtion con- sisting of a wcuum tube IT1:m(1 :I tran- sistor, Q1, connected in series. We can apply much the same approach as we did And if: HYBRID CASCODE for the cascode vacuum-tihe amplifier. Let AMPLIFIER us assume the base to emitter junction of (2, to Ix forward biased. The collector cur- rent of Q, becomes the plate current oi ~\llichis usually the case; then, V,. Thcrcfore, any clmnge occrirring at the 1):ise of Q1 is reflected as a change in Figure 13. A typical hybrid cascode amplifie plate current in V,. using a transistor and a vacuum tube. RL= LOAD RESISTANCE RL= LOAD RESISTANCE

# RE= EXTERNAL EMITTER RESISTANCE +& REFER PART I *THE TRANSISTOR SUBSCRIPTS WAND (2) ARE ASSOCIATED WITH Vt AND Vg

Figure 16 8 We recall (Part 1, The Transistor Am- plifier, Eq. 10) that the input impetlance we see looking into tlie base of a transistor 'the common-emitter configuration is:

I, also p = -- I I, therefore substituting equation (37) in equation (36) now the collector current Q1 becon~es tlie plate current of V1. Then,

El,, = I, (RE + Rt) since I,, = I, (39) also El, = -1,Rr. (23) TRANSFER CHARACTERISTIC TRANSFER CHARACTERISTIC OF TYPE 6DJ8 / ECC88 OF TYPE 2N408(PNP) and since Figure 14. The transfer characteristic curves of a vacuum tube (6DJ8) and a PNP trons~stor(2N408), illur El, rotina the basic similaritv between vacuum tubes and transistors. A(,) (stage) = - - I< i ,, then from equations (23) arid (39)

II>R1. A(") (stage) = - I!) (RE + Rt)

If the vacuum tube is not a triode but sonle other niultigrid tube such as a pen- Rk RE GAIN=A(v)= - GAIN=A(,)= - tode, the gain equation will have to be Rk+rk RE+R~ ~nultiplied by the plate efficiency factor (?I). ROUT = rk in parallel ROUT = (R+ +$) in parallel with Rk with RE The same remarlcs concerning tlie out- put impetlance of tlie vacr~um-tul)c cxs- Figure 15. The onalogy between the cathode follower (grounded plate) and the emitter follower (the common collector) in terms of "gain" and output impedances of both devices. code amplifier can I)e applietl to tlie liyl~~itl counterpart.

\Ve have sllo\vn tI1;it the gain of :: linear if \vc cl~oscto iqnorc tlie input iui1)ed;incei amplifier, tr;uisistor or v;lcuwn tulq is ;I of Imth tle~ices ratio of impctl:r~ices. \\'e can, of course, derive the gain equations for 11otll devices Figure 16 suniniarizes the results of our in terms of mutual contluctmce. Tn i:ict. analysis of the grounded cathode, grounded if we compare the transfer curves of 11otl1 grid, and grounded plate atnplifiers. devices, Figure 14, we see ;I striking simi- larity. Vnir and Ex can 11c tliougI1t of in It is not surprising we sometimes find the same terms and in like minner I,, and ourselves explaining one device in terms I, perform itlcntical functions. Our :~n;ily- of another. Nature has a charming way sis oi 1~1thdevices has sIio\v11 t1l;it this of making no st things interdependent upon fact is not coincidence. one another. Recognize this fact and most tasks become a little easier. It is not unrensonable to s:iy that \vhen \ve comp:irc the cathotlc-follo\ver (grountled- te) against the common-collector con- ,UI-ntion, Figure 1.5, xve c;ui tl~inlcof hot11 devices :is I~eing identical in operation- differing only in concept. The s:me argrl- ment can be put iorlvard about the com- PART 3 A DC ANALYSIS OF A TYPICAL TEKTRONIX HYBRID CIRCUIT

As a typical example of a Telitronix, Inc. hybrid circuit on which to demonstrate our DC analysis, we have chosen the vertical amplifier of a Type 545B . This circuit is representative of the hybrid circuit one encounters so often in electronic instru~lxntationtoday.

The Type 5453 vertical amplifier is a hybrid push-pull amplifier operating in a class A mode. It incorporates a few extra circuits such as trigger pick-off amplifiers necessary to acco~nplisl~its function, but, l~asically it is a hybrid push-pull amplifier.

Figure 18. The circuit which will determine the DC emitter currents for either Q514 or Q524. (A)-Thl actual circuit as shown in Figure 17. (B)-The equivalent DC circuit considering R517 or two througl To hegin our analysis of the amplifier, which the individual emitter currents will flow. the first thing we must do is select a portion of the amplifier circuit \vhich will give us the information necessary for us to make \Ve are now al~leto calculate the emitter Our nest stel) is to find the value of KI: our first calculation. \\'c ;~rcgoing to current of eitl~er(2.514 or Q524. The DC- \\.e tnust lino\v this value in order to cnlcu- analyze the \vhole circuit so we can choose enlitter current will flow tl~rough R51.5 or late gain. You will recall that I<,: will be our point of entry. The input circuit is as R516 and into R517 to grou~~d.Since the t11nt impetlance through which the signal good a point as any. Bear in mind that, emitter currents of (2513- :~nd Q524 botli current will flow to the AC ground. Let for our purpose, tliis is not the only point of pass tl~rough R517, we may think of R517 us t;ke :~notherlook at the resisti\~cnet\vorli entry. Any point on the circuit \vhich ~vill king mxle up of t\vo resistors, each of 1)etween the emitters of Q514 and Qj3 give us useful inform:\tion \voultl do. 2.6 lcn in value, in vi11ich tlie individual emit- The signal currents flowing in tliis circi. ter currents will flow, refer to Figure 18: will Ix cqu:~l antl opposite at t\vo points, Therefore, refer to Figure 19. These points are virtu:tl AC-ground points ; therefore, the impetlance A quiescent DC voltage of $67 volts is seen by tlie signal current froin the emitters the nominal voltage at the output of the of 52514 or (2524 \\ill be the p:~rallel com- plug-in aniplificrs used in the Type 545B hination of 153 12 and 27 fi or approximately oscilloscope. This voltage appears at term- 23 11 to the AC ground points. Hence, RI: ids 1 and 3 of J11 in Figure 17, and thus, for Q514 or Q524 will he 23 f? at the grids of V494A and V494B, a 6DJ8 dual triode. The input cathode follo\ver (V494 A & B) has a bias of about 4 volts; \\'e can now calculate the value of r,., tlie therefore, hot11 will be at +71 volts. dynamic-emitter resistance, The base voltage of Q514 and Q524 is then fixed at 71 volts. This sets the emitter of (2514 and Q524 at one junction drop more negative (they are 110th NPN R5l5 VIRTUAL AC transistors) than the base. Therefore, the 27 GROUND POINT: voltage at the emitter of Q514 and (1524 is 70.5 volts. T500 is a small toroidal trans- former used for high-frequency cominon- mode rejection. The DC BALANCE Con- to this vie can add our constant, R,, of say, trol, R495, sets tlie quiescent condition. \Ve 4 11. \Ve recall that: mean by this that the trace is centered.

We have tnatle certain assumptions allout the hias of a vacuum tube and the hase-to- therefore : emitter voltage drop of a transistor. This is quite justifiable since we Icnow what function the device performs. One helpful hint about transistors is that you can expect a base-to-emitter volkge drop of almut 0.5 or :tpprosimately 5 21. \\'e have now estall- Figure 19. The location of the virtuol AC ground to 0.6 volts for a silicon transistor and lislied the value of the emitter current and paints between the emitters of Q514 and Q524. about 0.2 volts for a germanium transistor. the value of Rt for Q514 antl 52524. I

We have now calculated from this part of There is one point we should make clear The trigger picli-off amplifier Q523 is the circuit all of the information we need here. We have assumed a value of 49 one part of a transistor cascode amplifier. to progress further into the circuit. Let us for R, which you will recall is equal to The input stage is Q514 and Q524. Nor~i~all turn our attention to the circuit around Rb- . R, can vary from between 2 R to the gain of a transistor cascotle amplifier Q513 and Q523. The first thing we notice B the ratio of RI. to I&: + Rt. The gain in tl is that the base of Q513 and Q523 are tied 24 R depending upon the type of transistor case must I,e multiplied hy 0.5 for the fol- together at an AC-ground point. You will (refer to Part 1, "The Transistor Ampli- lowing reason. The signal current is equally recall that the impedance we see looking fier" SERVICE SCOPE #42, February divided at the plate of V514B, half of the into the emitter of the common-base con- 1967). This is one of those few times we signal current will flo~vthrough the figuration is Rt. In order to calculate should be really a more specific about line iml)etiance (93 a) and the other hali Rt we must, of course, calc~~later, and add assuming a value of R,. The sum of the through R526 and finally through the load our constant for R, of 4R; r, will be a impetlances 5.68 R and 90.9 R should be equal impedance of Q523. The load iuipetlance will function of the actual value of current flow- to 93 R since our delay line is a 186 f2 balanc- be that impetlance ~vliichis connected to the ing into the emitter. 27 niillianips has been ed line. Therefore, we have a difference of AC ground. The collector of Q523 is con- set in the emitter circuit of Q514 and Q524; 3.58 R between the theoretical value and the nected to the base of Q543. The im~)etlnnce but not all of this current will flow into the calculated value, or an error of approxi- we see loolting into the base of 52543 is mately 3.770. This error has been due in part to our presupposed value of R, to be 4R. Such an error could not be tolerated in design work but it is acceptable here for our purpose of DC analysis. Bear this limita- tion in mind when you apply this analysis. If \ve choose to neglect the input circuit of the trigger amplifier we see that R1: in TO AC this case is R547 6.5 IiR. A beta of 50 is a GROUND POINT close figure to use for Q543, and since RI: 3.01 K There is another point we must clear up. > > R, then, What is the load impedance of the hybrid cascode amplifier Q514, V514A or Q524, V514B? Clearly it will be that impedance or impedances connected from the plate of V514A or V514B to the AC ground. We are using a balanced delay line of 186 R, (93 R to a side), referenced to the AC ground. There- TO EMITTER Q523 VIA R527 (3.01 K) fore, the delay line impedance (93 9) must shunt R511 in series with Rt (or R526 in series with R,) making an effective load impctlance of approximately 47 R in the plate Figure 20. The DC current paths in the emitter circuit of V514A or V514B. We now have circuit of Q513 or (2523. all the necessary information to calculate This impedance shunts R54 (75 kll) antl the gain to this point. 1228 a 1.5 IiSl ~vire-woundresistor. \\'e may emitter of Q513 and Q523. 11.5 milliarnps then, for all practical purposes, consider will flow through R510 and R527, refer 1-528 the collector load resistance (RL); to Figure 20. The actual value of current therefore. into Q513 or Q523 will be 15.5 milliarnps. Therefore, the impedance (Rt) we see look- ing into the emitter of Q513 and Q523 will be

Q534 is the heam-indicator amplifier. Its function is to drive t\vo lamps situated above the CRT on the front panel of the oscilloscope. These indicate the posi- tion of tlle trace in a vertical direction. In the quiescent condition the voltage at the This impedance of 5.68R plus R511 or junction of R535 and R536 is 287 volts. Bof' R526 (90.9 R) constitutes part of the load Q523 is the trigger pick-off amplifier and indicator neons, I3538 and B539, have , impetlance of the hybrid cascode atiiplif ier Q543 is an emitter follo\ver providing isola- volts across them, not enough voltage to Q514, V514A or Q524, XT514B and the neces- tion between the vertical amplifier antl the strike either neon. (This type of neon has a sary niatching impetlruncc for the delay line. trigger circuits. striking voltage in excess of 68 volts.) When we apply a negative signal to the Therefore, the voltage at the junction of a step function. The DC SHIFT control vertical input of the osc~lloscope,the base of R535 and R536 rises towards 350 volts is adjusted for the best dynamic tliermal Q524 is dr~vennegative and ilie base of striking neon 3539 which indicates trace compensation, typically about 1% tilt. "I4 moves in a positlve d~rection by a has shifted down. ~laramount. Tlierefore, tlie current \Ve will now analyze the output circuits tlirougli R530 tlecrcases and the current R513 and R523 and the DC SHIFT con- to the riglit of the delay line, refer to Figure through R507 Increases The voltage at the trol R502 are thermal-con~per~sationnet- 17. The first thing we must do is to cal- emitter of 12531 increases and the voltage :~t works associated \vitll Q514 and (2524. culate the voltage at tlie base of Q594 or the base of Q534 decreases As a result, the The thermal time constants are long and Q581. TIie voltage at the junction of R532 hase-to-em~ttcr junction of Q534 beco~nes the visible result appears on the CRT dis- and R533 (174 volts) will set the hase volt- reverse b~asedanti Q534 ceases to conduct play as a DC shift in trace position after age of Q513 and (2523. Assuming a junction drop of 0.5 volt tlie voltage at tlie emitter of (2513 and Q523 will be 173.5 volts. The cur- rent through R511 and R526 is 27 milliamps, hence the voltage drop across these resistors will be

+ 350 V GRID V594 T z 2.5 volts

therefore, the voltage at tlie plate of V514A and V514B is 17.8K

GRID 173.5 - 2.5 = 171 volts. V584

902 This 171 volts is directly coupled to the base 1 E2=R2of Q594 and (2584 via tlie delay line. The voltage at tlie emitter of both Q594 anti (2584 is then 170.5 volts. We will now calculate + IOOV ('42) the current flowing into the eniitter of Q594 or Q584. Figure 21 shows a step-by-step approacli in solving this problem. TIie sim- plest approach is to use Th6venin's Theorem to simplify the resistive network R569, R570, R571 and R572. The result is we have a V,, of +I00 volts and a Ztk, of 9OOQ to the junction of R574 and R576. Therefore, Zth =900 looking from the emitter of either (2594 or Q584 we see an impedance of 13.3 n in series with 1800 Q to +I00 volts.

voc "IOOV + T

(VI -V2) R2 Voc = V2 + RI +R2 we now calculate r,

26 26 r*r-r- I,: 39 Vo, I00 VOLTS

and to this we add our constant R, of 4 n; tlieref ore,

re 21. Illustrating the use of ThBvenin's Theorem to simplify a network consisting of a voltage sour, ,.,d a resistive network. (A)-The network whose Thevenin equivalent is to be determined. (B)-Determi ing the equivalent source impedance (Zth) and the equivalent voltage source (V,,). (C)-The Theven equivalent network of (A) connected to the junction of R574 and R576. (Dl-The equivalent circuit co sidering Ztt, as two resistors through which the individuol emitter currents will flow. We have only one point in this circuit (a A beta of 75 for tliis type of transistor is value of Iii,, is \vitIiin practical limits. Figure virtual AC ground point) at \vliicli tlie sig- ;I close figure to use for prxtic;il purposes. 22 sho\vs ;i progressive l~rcalitlown of tliis nal currents will he equal and opposite. That Tllerefore, 11etworIc. point is the junction of R574 antl R576 (13.3R resistors). Tliis fact sets R1.: at I?!,, = 75 (13.3 + 4.7) R's 13.3 R. Tlie purpose of the RC network to Tliis net\vorlc will induce :1 loss bet\vecn ,. the right of R574, R576 is to compensate tlie - man the two stags. I lie signal is reduced in high frequencies. amplitude 1)y a f;ictor of 0.64 11ec:~useof the The value of lii,, is part of a resistive net- voltage tlividcr tict\vorli coiisisting of 57.6 St Tlie input impcdmce we see looking into work whicli will terminate tlie delay line in ;end the p;ir:illel combination of 100 R, the base of Q594 or (2584 is its correct impetl;cncc. Tllerefore, lxfore ~ve 2800 11, and the input impetl:uncc into Q594 or Ri,, = B (K: + Rt) (10) Icnve this section uc must check to see if our Q.584.

Tlie gtin of the output stage is HIGH FREQUENCY NETWORK

t! J! 594 or 584

IMPEDANCE I - AC GROUND

You recall that tlie gain equation of a liytlrid cascode ;mplifier (refer part 2, "Tlic Vacuum Tube Amplifier," Service Sr #43, .4pril 1967) must he multiplied by .. . pI:~te efficieticy factor (7) if the vacurlm tube is not x triode. The plate efficiency factor (7) normally varies f so111 bet~veen0.7 to 0.9. In tliis case (*) is approximately 0.9 - 0.88 to be exact. So finally,

NET IMPEDANCE TO AC GROUND -B

The gain of the co~iipletc Type 545B vertical amplifier is

A(,) (total) = 54.9 X 1.68 X 0.64

Tliis brings to a close tliis series of three AC GROUND = 1350n articles dealing \\.it11 a practical approach to tr;uisistor antl vacuutii-tuhe amplifiers. -C This appro;ccli has Ivxn offered as a direct method of troul,le shooting and untlersta~i(l- ing circuits. Tlicrc are limitations as tc Figure 22. Illustrating a step-by-step approach for analyzing the circuit between the output of the del a1)plic:ltion ;is \ve have seen. I-Iowevcr, tlic~e line and the base of Q594 or Q584. (A)-The circuit between the output of the delay line and the bc limitntions (lo not impair tlie practical of Q594 or Q584. (8)-The net impedance of the circuit in (A) to the AC ground between the output

the delay line and the base of Q594 or (2584. (C)-The input impedonce into Q594 or Q584 and the I ;lppro:~cli \ve must apply to our everyday impedance, os shown in (B), providing the terminating impedonce for the delay line. mainten;ince antl trouble shooting problems. LIST OF SYMBOLS

~Eout Base spreading resistance (Tee The "Transresistance" resistance Voltage gain defined as -- AE,,, Equivalent) (re $. Rr) Collector resistance (Tee Equiva- lent) Average or quiescent value of plate Vbb Base voltage voltage Emitter resistance (Tee Equiva- lent) v,, Supply voltage Plate supply D-C voltage External Emitter resistance (refer vcc Collector to emitter voltage Average or quiescent value of grid to text) voltage The equivalent resistance between (alpha) The total forward current Effective or maximum value of the signal source and the emitter gain of the transistor as viewed varying component of grid voltage terminal of the transistor in the from an external circuit. Normally common base configuration Average or quiescent value of cath- defined as the ratio of I& ode voltage Dynamic emitter resistance (beta) The small signal current gain Effective or maximum value of Cathode resistance (refer to text) (delta) The change in the variable varying component of plate voltage The impedance seen looking into with which it is associated the cathode of a vacuum tube and (eta) Plate efficiency factor (refer Mutual conductance defined as to text) rp + RL 1 --- - (if rp > > Rr.) (mu) Amplification factor P+1 gm Base current Load resistance Collector current Dynamic plate resistance Emitter current Cathode current Plate current For additional information or for Order Placement contact:

P. 0. BOX 500 . BEAVERTON, OREGON 97005 . Phone: (Area Code 503) 644-0161 . Te!ex: 036-691 Tektronix, ~nc./ rwx: 503-291-6805 . Cable: TEKrRoNlx . OVERSEAS DlsrRlBUroRs IN OVER 30 CouNrRlEs TEKTRONIX FIELD OFFICES in principal cities in United Slates. Consult Directory Tektronix Australia, Ltd., Adelaide; Melbourne; Sydney . Tektronix Canado Ltd., Montreal; Ottawo; Toronto; Vancouver Tektronix International A.G., Zug, Switzerland . Tektronix Ltd., Guernsey, C.I. . Tektronix U.K. Ltd., Harpenden, Herts

X-2355 Printed in U.S.A. KLW