Text preview for : 1965SC.pdf part of Tektronix 1965SC Tektronix publikacje 1965SC.pdf

Back to : 1965SC.pdf | Home

Patrick B. G e e
Tektronix Field Maintenance Engineer

ferent clinracteristics. Your pri~icipal objec-
tive should be to first diagnose the ills by
knowing tlie sj~rnpto~iis, select the most suit-
able re~netly,and t l ~ e ,perform tlie operation
Editor's N o t e s
you have selected.
Tlze basic concepts rcviczwcd in this
Let's review some basic s;u~iiplitigconcepts
article apply t o all s a i n p l k g ifzstrzc-
witli the intent of learning what c1iar;icteris-
nlci~ts T h e iuf o r i ~ a t i o i ~ ,
however, as
tics a r e c1i:itigcd by each :ltljustmcnt within
it i s preseiztcd lzerc w a s developed
the s:impling "1ie;d". First of all we sliould
aroui~d tlzc calibratiofz of tlze T e f i -
Iiave a good u n d e r s t a ~ l i ~ i about ~ t l i ~ ~ f i l i t ~ g
tro~zix T y p e 4 S 1 Dual-Trace S a m
cfjicirt~cy which is ;I measure of signal
plilzg U n i t I t i s directcd principally
transfer across the bridge diodes sampling
toiuard tlzosc zulzo, .rulzef~exposed to
g:\te. Consider tlle di;lgram shown in Figure
samplifzg t c c l z ~ ~ i q u cfeel the w e d for
1. O u r purpose in ope~iing tlie s;uiipling
a bit illore sltpport By devclopifzg a
gate is to permit the s:cmpliiig c;lpacitance
fiillci. zt~ldcrstaiditzg of tlzese impor-
tatlt feattires of fzwzdameiztal coi~ceriz, (C,) to "see" the input sign:xl for :t small
period of tinie, tlie duration of the snrnple
the attt/zor lzopes to supp1y this sup-
being a limiting f : ~ t o rof system risetime.
fort am' t o dispel tlzc ~ced1e.s.s fear of
(Instrument risetinie can be no faster t1i;in
sairlplii~g that .scci~~.sto lzovcr i n tlze
the length of time the sampliiig gate is
iuilzds of .soiue
open.) \\'e Imow that it invariably takes
This article has been prep;ired f o r those some time to fully c1r:irge ;i capacitor 11c-
involved in the c:~libratioti o i the Telitronis c;iuse the source :111d current 1mtl1 Iin\.e
Sariipling Units, witli the T y p c I S 1 Dual- imped:uice. T h e pre-atiiplifier Input cnpnc-
Trace Sampling Unit being used as an e s - itnnce ( f r e q u e n t l ~cnllctl the s:~rnpli~lgc;lp:lc-
ample. I t is intended to dissolve :L few or- itance) in tlic Typc 4S1 ~ 1 1 1cllarge to
dinary misgivings a l ~ o u t approaching the
unit and to outline an orderly and effective TABLE I
method of system diagnosis :uid tre:ltiiient. Escellcnt p e r f o r m a ~ ~ cshould not he e s -
You slioultl malie nn c f i o r t to tliorouglil~ pected from r;\ntlom adji~stments. Ii;ltlicr,
understmd wli:it each adjustment acco~ii- :ui ortlerly and sj.stematic :~p~)roncli must be
plislies. Once you attain this ol~jective,you talien to rcstorc the Type 4S1 to its proper
will no lo11gcr need to rely on detailed c1i;iracteristics. Adjustnient is neitlicr :i tlii-
instructions to calilxlle tlie instrument. You Licult 110s :ln estrciiiely simple thing to do.
sliould find it possil~le t o periol-m all the i i c \ ~:djustmcnls, 11cc:luse they Iinw :111
necessary :~djustmcnts on a Type 4S1 in a effect on ~evcr;il different cliarnctcristics
very few minutes. I'eriorming all the (all o i \\.Iiicli we \visli to lioltl \vitliin spec-
clieclts that insure tlie inslru~iietit meets ified limits), confound the rec:1libr:1tion.
original specific:itions may, llowever, t;iIi~ Tlic c1i;trt (see T n h k 1 ) slionrs the :djust- Figure 1 . Schematic of o simplified sampling.
an hour o r more. lncnts thnt have an c i i e c t on sevcml dif- bridge gate.
o n l y about 25% of the difference in voltage 6 l l e gate with only one sample. T h e method Obviously then, anything tliat we do within
a c r o s s the sampling gate in 0.35 nanosec- used here is to amplify the change in volt- the sampling loop tliat changes either Sam-
o n d s . This percentage is referred to as the age on the sampling capacitance and add the pling eff~ciencyo r gaiu within the loop will
sampling ef f iciencp. amplifier voltage to this capacitance be- a150 change dot transient response. I n other
words, dot t r a m i e t ~ t response is a fxnction
of both s o ~ i ~ p l i n rfficirncy nnd loop gain
Suppose that the product of sampling ef-
ficiency :uid the amplified feedback signal
were to equal more than unity. Our pres-
entation would tlien appear to h w e over-
shoot and/or ringing as sliown in Figure
3. This is just a s undesirable a s the rolled-
off presentation sllown in Figure 2.
T h e four-diode sampling gate perfortiis
a few functions which require further ex-
planation. During quiescent conditions the
- Kate is closed so that the signal cannot pass
Figure 2. Waveform of the exponentiol in- volts on tlie first sample. Between samples through. T o do this, we back-bias the gate
crease in the sampling-capacitance charge we could a n ~ p l i f y the charge with an am- with a positive antl negative dc voltage of
with each successive sample, when very few plifier having a gain of four antl feed hack approximately two v o I t s . T 11 e dynamic
samples/division are taken.
this one-volt signal level to the sampling range of the gate is limited by the tiiagni-
Since tliis capacit:mce will not be ~11s- capacitance f o r a period of time that permits tutle of tliis 11 01 d o f f b i:i s ( B R I D G E
charged between samples, we would expect full transfer of the charge. \\re then end V O L T S ) ; a signal greater tlian two volts
t h e charge to increase exponentially u i t h up with the required one volt across the might overcome the holdof f I ~ i a sant1 im-
each succe5sive sample a s sliown in F ~ g u r c s:unpling cnpacitance. Now we can cliange properly c1i:u-ge the sampling capacitance.
2. O u r system uoultl tlien reconstruct a the number of samples per division and the A trigger pulse from the timing unit initi-
pulse with severe rolloff even from an infi- trmsient response of the observed waveform ates tlie generation of tlie strobe pulse (to
open tlie catiipling gate) and the memory
gate pulse (to open the 1iwnory gate). T h e
amplitude of the narrow strobe pulse riiust
be sufficient to rise above tlie holdoff
bias for a period of time T , thus forward
biasing tlie bridge diode gate :is sliown in
Figure -5. A n increase of strobe ampli-
tude will usually cause an increase in sam-
pling efficiency because the sampling ca-
pacitance has longer exposure to the in-
put signal and therefore can charge to a

Figure 3. Tektronix slide-back, feed-bock sampling system.

nitely fast step function if very few samples sliould r e ~ n a i n tile same. \\.e could say,
per division were taken. T h e rolloff would then, tliat our "dot transient response" is
become less obvious, of course, if more correct since we have a gain of exactly one
samples per division of 11orizont;~ldeflection tl~rougli the entire loop wlien referred to Fig. 5. The narrow strobe pulse rises above
holdoff bias for a period of time "T" to for-
were talcen. F o r example, if 10 samples the input signal. (Remember, though, tliat
ward bias the bridge-diode gate.
were required to fully charge the sampling this required n goill of fozo. when referred
ca~acitance,the rolloff \voultl he evident f o r to the charge on the sampling capacitance.) higher voltage. Also, a higher strobe am-
plitude will cause the diodes to exhibit a
lower impedance during tlie sampling inter-
val. T h e gain required through the am-
plifiers and feetlbaclc attenuators to yield
:L loop gain of unity (correct dot transient
response) is the reciprocal of sampliug e f -
ficiency, so w e would need to reduce loop
gain to compensate f o r an increase in
sampling efficiency if w e were t o niain-
tain proper dot transient response. Note
that a reduction of B R I D G E V O L T S
(keeping strobe amplitude constant) could
cause a s i ~ni I a I- c l i a n g e i n sampling
cause the sampling capacitance to become fied feed-back being greater than unity. efficiency
SJi-YCu4.?r'&o 4 e 'THE C.?.?OX V d i r f i Z - B
A few words are in order concerning the to the pre-amplifier input capacitance-one VERTICAL POSI- Midrange (dot to 12
generation of strobe pulses. A trigger pulse that will track with the ac amplifier at- TIONING O'CIOCI~)
from the 5T1A timlng unit causes the nor- tenuator. This will increase the feedback
mally forward-biased snap-off diode to bc- applied to the sampling capacitance as the SMOOTHING Normal (Maximum
come reverse biased by a reverse current of ac nniplifier signal is decreased (as referred loop gain)
high and relatively constant amplitude. A to the signal applied to the 451 input con-
peculiar characteristic of the snap-off diode nector) witli less sensitive settings. R, and nc OFFSET Adjust for zero volts
is that this large reverse current ends very R, make up tlie second attenuator. (Atten- t 100 niv at the
abruptly (within a few picoseconds) and the uation is reduced liere when it is increased DC OFFSET
snap-off diode becomes a very high imped- between amplifiers wit11 1~0th attenuators M O N I T O R jack
ance. The reverse current that was flowing operated by tlie same control laiol~). The
down the 50-fi shorted transmission (clip) resistor divider ratio of tliis pair determines \Wth a free-running sweep, both traces
line in trying to continue to flow, produces the hasic calibration of the sampling loop. should be well witliin the central graticule
a voltage pulse of short duration that ovcr- area of a properly adjusted instrument. Se-
Another diode gate precedes tlie Meniory
comes the back bias on the sampling gate verity of imbalance is often indicated in this
stage. \Vllen the fast, narrow strobe pulse
and causes the diodes to conduct. When all display and your observations here may help
is generated, a relatively wide (250-350
the bridge diodes are conducting, they repre- in the diagnosis. If the presentation looks
tianosecond) pulse is also generated to open
sent a low impedance path for the input sig- otlier tlian normal, first perform steps 3
the nieniory gate. The paramount functions
nal to get t o the input preamplifier. When and 4 of the recalibration guide which fol-
of the memory gate circuits are t o : ( 1 )
the voltage pulse is reflected (after about lows this article and then start back with
control the in-phase feedback to the sani-
0.35 nanoseconds) due to current traveling stop number 1.
pling capacitance and prevent the memory
in the shorted clip-line, the sanipling gate is
from responding to this regenerative feed- Several methods, each having its own
returned to its reverse-bias condition thus
b x k signal, (2) insure maximum coupling merits, may be used to show dot transient
locking out the input signal once again. The
of the ainplified error signal to C?, and (3) response error. A most useful nietliod is
combined snap-of f diode and clip-line action
limit memory capacitor discharge between to apply a step-function to the input and
produces a very fast rising and falling pulse
samples. (Lealiage of the cliarge in tliis use a sweep speed that will display no niore
of a very short controlled duration. A n -
capacitor causes vertical deflection of the than two or tliree sanlples 011 the leading
plitutie of reverse current in the clip-line is
dots l~et\veen saniples anti is called Meniory edge of the pulse (low vertical dot density)
determined by the stored cliarge in the diode
Slash.) It limits the maximum permissible at 100 or niore samples per division. Should
which is a function of forward S N A P -
time between sanil~les for a useful display. the p d s e shape or transient response change
O F F C U R R E N T . Reverse current must
Tliis leakage is caused by Memory A n - when switcliing from 100 to 10 or fewer
be sufficient in ulagnitude so that the volt-
plifier grid current or diode gate leakage. samples, then dot transient response is not
age created while it travels in the 5 0 4
A cursory analysis of the system as shown correct Quite often in using only 10 sani-
clip line is more tlian enough to overcome
in Figure 3 reveals tliat the following con- ples/div an important part of the trace may
the holdoff bias on the sampling gate.
trols all have a direct effect on dot tran- be missincr and the overshoot or undershoot
Let's refer acain to Figure 3 and review sient response : tliat appeared witli 100 saniples/div will not
some of the primary objectives here which 1. Those that control scl~i~pling f f i c i r n c y
r be displayed because it occured between
are : (1) charge C, to the amplitude of the are dots in the presentation. Therefore, when
input sigual as much as possible during the operating at few saniples/div you may need
useable period of the strobe pulse to increase a. S N A P - O F F CURREKT-common to to relocate the dots along different portions
sampling ef ficiencp, (2) feed back an am- hot11 sampling gates of the trace or "slide" them back and forth
plified version of this signal between s a n - 1,. B R I D G E VOLTS-one for eacli sam- to simulate a solid trace by rotating either
ples to charge C, to the full level of the pling gate tlie T I M E P O S I T I O N or V A R I A B L E
input signal, (3) simultaneously cliarge C.. T I M E / C M control. ( T h e slow sweep speed
to a value proportional to the input signal 2. Those tli;~t control loop gain a r e required for low vertical dot density usually
level and permit C2 to retain this charge places the beginning of the pulse towards
a. .AC A M P L I F I E R GAIN--one for
long enough for us to observe low rep-rate tlie left edge of the crt. Using the V A R I -
each sampling gate
signals. A B L E T I M E / C M control is generally more
11. M E M O R Y G A T E WIDTI-I-coni- desirable for this situation since it moves
The voltage on the Meniory Capacitor is rnon to I>otIi memories the trace to the right, towards the center
proportional to the input signal and is used of the screen.)
to drive the scope's vertical atnplifier. T o c. SMOOTHING-a front panel control
deflect the dot a given distance with a for eacli ac amplifier .AnotIier method requires a generator of
larger signal at the input requires attenua- the mercury-pulser variety (Tektronix Type
tion of the larger signal before it is applied The primary purpose of tlie SMOOTI-1- 109 or Type 110) \vith a small cliarge line
I N G control is to reduce random noise by on one side of the s\vitch and no cliarge line
to the Memory. In otlier words, the Memol-y
output signal will normally always be pro- reducing gain of the ac amplifier. Since on the other side Here the sampling gate
tliis is within the feedlxick loop, it neces- is opening on the two inputs alternately
portional to tlie deflection it causes. Stray
sarily follows that dot transient response Tlic sampling capacitance most of the time
capacitance and other factors proliibit using
will be effected corresponding to the amount must alternately charge from the amplitude
;I switclied altenuator at the input connector
of smoothing used, but may not be apparent extremes between the voltage at the top of
f o r reducing the deflection sensitivity. I t
when using lots of s;lmples. the pulse input ant1 the zero volts from the
is more feasil~leto use an attenuator at tlie
pre-:uiiplifier output to limit the signal Your preparation for recalibration and/ otlier input. Response witli eacli sample is
coupled to the high gain a c amplifier and or repair sliould include tlie following ad- manifested in the display. Proper D T R
also prevent overdriving this Stage. But ditionnl presets on the Type 4S1 : (dot transient response) would give a pres-
we niust maintain loop gain close to unity. entation tliat should look like Figure 6a.
Tliis requires :I second attenuator in the MV/CM S\\IITCH 200 Low loop gain \vould give a presentation
feedback path from the Memory Capacitor V.4R1.4BLE Gilibrated tliat sliould look like Figure 6b, and esces-
Fig. 6. Waveforms of: ( a ) correct DTR, ( b ) low loop gain, (c) excessive loop gain, using a small 4. *('j'lst BALANCE f o r
charge line on one side of the switch and none on the other.. 110 trace shift while rotating S M O O T H -
I NG-both channels.

when switching from N O R M A L to I N -
Figure 7. Woveforms of: ( a ) correct DTR, ( b ) low loop gain, (c) excessive loop gain, with the
5T1A set up for +
INTERNAL triggering ot o sweep speed of obout 2 NSECICM and using o TU-5 VERTEI) ( D C O F F S E T M U S T R E
Pulser/Adopter operated by a 2 5 - k c square wave from a Type 1 0 5 Squore-Wave Generator to
drive the Type 4S1.
T h i s cotnpletes the adjustments for the
sive loop gain would give a presentation simple process if previous calibration set- Type 4S1, leaving ouly a series of checlcs
that should look lilce Figure 6c. tings have not since been mis:djustetl. T h e tint should he performed to insure that the
L o w repetition rates inherent with mer- following method map be used to perform instrument is functioning properly. T h e
cury-pulsers a r e sufficiently atlnoying to routine recalil~ration. T h i s is fzot ;I complete ti~o\t irnport:\nt considerations include :
w a r r a n t investigating other ways of obtain- rec:ilil~ration procedure, but should serve a s
a useful reference in conjunction with the :L RISETTME-less than 0.35 nano-
i n g a similar "twosies" type of display. One
regular rec:ilibrntion procetlure in the in- seconds computetl.
such way follows, but requires tlc internal
. .
triggcrln: : Using :I Type 5 T l A Timing struction manuals. I). NOISE-less t11:in I mv (consider
Plug-111 Unit set up f o r + INTERNAL 1. Adjust M E M O R Y G A T E WIDTI-I for 90% of the (lots).
triggering at a sweep speed of ahout 2 marinium loop gain (LC., maximum over- c. B A S E L I N E SHIFT-less than 3 mv
i\'SIle:di~lg edge of :I pulse from a Telctronix sient response). kc rep-rates. ( T h i s is a shift of
Tjyx TI!-5 I'ulser operated 11); :t 25 kc N O T E : Before adjusting S N A 1 ' - O F F the tlc reference level o r base - line
squ:lre wive f r o m the Telctrotiir Type 105 C U R R E N T o r B R I D G E V O L T S , first wit11 cI~;u~ges rep-rate. I t may come
Squ:irc \fravc Generator. S~vitcllingthe 4S1 determine which :itljustments ~lcetl to be Trotn several sources including improp-
triggering sn.itch from ;LC to tlc trigger nlatle hy application of the following er ;djusttiients, ant1 is usually greatest
coupling shoultl produce n display similar concepts : l~etween90 l;c to 100 Ice. Scaling drift
t o those shown in Figure 7. I-Iere the trig- is cl~eclcctl by o1)serving a trace with
2. Clrrcli I I T R o n both cll:im~els.
g e r circuit is :dtern:ltcly respondi~lgto the n o signal applied and triggering the
a. If the snliie D T R error exists on both
lending edge : w l pulse top. Triggering on s u w p from I 0 cps to 100 kc using a
cll:~mels, adjust SNA1'-OFF CLJR-
t h e pulse top occurs 1)ec:msc the pulse top Type 111 P u l s e G e n e r a t o r or
R E N T for correct I I T R on 11ot11
is still more positive than the THRESI-T- ccluiv;ilcnt.)
OLT) setti~lg after trigger recovery takes
pl:tce 111:il;ing the T ~ y e T I A ready to trig-
11. If Cii:iiinel .A D T R is poor :mtl Ch:u- tl M E M O R Y SLASH-less than % cm
nel I D T R is gootl, :idjust Channel
3 vertical trace slash a t 10 cps.
ger ;\gain ~vllenevcr the T H R E S H O L D
A B I I I D G E V O L T S for proper
level is esceetletl. T'ulse :~mplitutlcafter the c. O V E R S H O O T o r U N D E I I -
n e s t recover>, cycle \vill I)c 11clow tile D T I I on Channel A.
S1 IOOT-3% 1li:1ximu1ii.
T I - I I I I 3 H O L I 3 levcl n.llic11 \\.ill nrevent c. If Cl1:umel is poor ; u ~ d Cl~:ul~iel A
is gootl, adjust Channel 13 RRT1)GII f. 1jOT T R A N S I E N T RESPOKSE-
the trigger cil-cuit from respondi~ig until correct for Imth positive :ind nega-
V O L T S for proper 1)1'I< on Cllan-
the next positive c s c u r s i o ~ ~ tllrougli the
T I - I R E S I I O T ~ I )Icvcl setting. T h e sampli~ig
1x1 P. , tive going signals of less than '/2 v *
(I. Tf Imth cl1:uniels e s h i l ~ i t D T I i crrol-s
cnp:icit;ince must tllcl-efore cll;irgc to the
in opposite directions (one sho\ving If risetime is :itlecluntc but noise and/or
pulse :~mplitutle cstl-emes during the first
f e w ce~ltimctcrs of tlis11l:~y with each suc- too much loop gain :mtl the other sc;lling tlriIt a r e e r c e s s i v e , decrease
sho\ving insufficient loop gain), per- I3III I ) G E V O L T S :ind rc:djust S S A P -
cessive s:unple.
form tlic follo\ving steps : O F F CVRIIEN'I' for proper clot transient
The cllnrt sllo\vn in T;d)le 1 is ;uiotller
( 1 ) Adjust BII1I)GE V O L T S on rcspo~lsc,tllcn repeat steps 2, 3 nnti 1 above.
11oth ch:ulncls to masi1i~u1i1 c1ocl;- hkilce sure that B R I D G E \'OLTS is at
wise positions. l e x t 2 volts :tljove and 11elow ground for
(2) Adjust S N A P - O F F Cl_~lilIES'Y !.our f i n d setting.
f o r proper 1 I T R on the c1l:ulnel
NOTI<-licfcr to your instruction m:wunl
o r r c c : ~ l i l ~ r ; ~ t iprocedure f o r otller cllcclcs
to Ijc performctl.
T Y P E 575 T R A K S I S T O R CURVE after extended periods of service) they may ( ) 3. Solder the new sleeving-covered wire
TRACER - NOISE O N HORIZON- be salvaged to give niarly more hours of and the white-violet wire to the ter-
T A L A N D VERTICAI. A T T E N U A T O R use. Indications of wear can be a noisy minals shown in Figure 1.
S\VITCHES motor, and/or excessive end play of the
motor shaft. (You sliould note here that
Under extreme environmental conditions,
a bent o r out of balance fan blade can
foreign riiaterial can build up on switcli
vibrate and give the appearance of a noisy
contacts and cause excessive electrical noise
motor. Check your fan blade before finally
T h i s noise can be particularly objection-
assessing the cause of noise.)
Tlie cause of noise o r shaft end play in
T h e application of a f h i ~ 7 film of Cramo- a 147-001 niotor is wear on tlie seven
l i ~ i cleaner and lubricant (Tektronix part wasliers shown in Figure 4. T o replace
number 006-197) will solve this problem. the washers shown in this exploded draw-
Usage of Cramolin will result in approxi- ing you will need:
mately 40 times improvement in reducing
Qty. Part # Figure 1 . Diagram showing solder terminals
noise ;md wear, over a dry switch. on relay K601.
2 each 210-980 steel w a s l ~ e r s
Cramolin should be applied with a srnall
4 each 210-982 I~eryllium washer5 ( ) Soltler the gray-red-red wire to the
artist-type camel-hair brush. Just a drop
placed on the b r u s l ~ and then applied to 1 e;d~ 210-981 fiber washer terminal 41own in Figure 1.
tlie switch contacts and rotor will give good ( ) 4. Soltler a 10" piece of #22 wli~te-
Thcse parts may be ordered tl~rough
results. A f t e r applic:ttion, rotate tlie switcli black wire and a 6" piece of wliite
your local Tektronix Field Engineer, Rep-
back and forth through its range several
resentative, Field Off ice o r Distributor. wire to the terminals shown in Figure
times. T h i s aids the cleaning anti lubrica-
T h e Mec11anic:d parts list in t l ~ cInstruc-
t i o ~ i action. Avoid the use of excessive
tion M:tnu;d for your instrument gives tlie
amounts of Cramolin. Anything more th:in Soltler the otlier end of tlie white-
Tektronix part numl~er f o r the fan niotor.
a thin film will only detract from the neat- black wire to C S H - I (locate in
W e remind you, the inform:ttion given here
ness of your work and will neither hasten Figure 2 )
applies only to instruments using fan motors
nor aid the c1e:ming and lubricating action.
part numbered 147-001.
C r a ~ n o l i n may he obtained through your
local Teldronix Field Engineer, Representa- T Y P E 661 S A M P L I N G O S C I L L O -
tivc, Field Office o r Distributor. SCOPE - DELAYED PULSE MODIFI-

FORCED-AIR VENTILATION - F A N H e r e is a do-it-yourself modification that
MOTOR SALVAGE will protect the Tunnel diode D992 (in the
Delayed Pulse circuit) f r o m excessive cur-
rent during the warm-up time of V694 and
V814. T h e modific:ition routes the current I CSG 9
supply through relay K6Ol until tlie instru- p o ~ ~ ~ , a n t )

ment is warmed up, at which time normal -7
supply current is restored. T h i s modiiica-
tion applies to Type 661 instruments serial
n u n ~ b e r s 101 through 2219.
Tlie followitig instructions sliould aid in 1 5
rewiring the relay : CSJ IZCU

I M l ' O R T A K T : Use silver-hearing solder
when soldering to cer;mic strips.
( ) 1. Unsolder from relay K601 :
Figure 2. Diagram showing layout of ceramic
strip terminals referred to in Delayed Pulse
Figure 4. Exploded drawing of fan motor, Modification.
part number 1 4 7 - 0 0 1 .
( ) sleeving-covered wire ( ) 6. Soltlcr the otlicr end of the wliite
Many Tektl-onix i ~ ~ s t r u m e n t eml)lo>iiig
s ( ) gray-red-red wire wire to CSD-19 (locate in Figure
forced-air ventilntioi~ usc tlic same type 2 ) . This completes t l ~ eniodification.
fan noto or. Te1;tronix p w t nuinl~er f o r ( ) 2. Replace t l ~ c slewing-covered wire
this motor is 147-001. \\'1ie11 tliese motors wit11 3 piece of wire and sleeving ( ) 7. Cliecli wiring Tor :iccuracy ;ind
begin to display sigus of \\.c;lr (norm;dly that is '/8" longer. c11;uige Interconnecting Sockets dia-
g r a m in tlie Type 661 Instruction and then set tlie Base Step Generators T o check f o r low emission in the crt,
Manual to agree witli Figure 3. S T E P S E L E C T O R control to 0.05 ma. remove the c.nlibrator signal from tlie crt-
This gives an I R drop between the gate cathode connector and reconnect the ground
and source terminals of the F E T of 0.5 strap Adjust the F O C U S and A S T I G -
volts per step. Tliis is sufficient to view M A T I S M controls for largest diameter
the complete family of curves from zero spot. \\'it11 the sweep turned off, adjust
to cutoff. tlie I N T E N S I T Y control to where the
de-focused spot on the crt face has a very
T Y P E 530, T Y P E 530A, T Y P E 540, slight halo. Remove the left-liand side panel
T Y P E 540A, T Y P E 540B, T Y P E 550, from tlie oscilloscope. Then, with the tip
T Y P E 585 A N D T Y P E 585A OSCILLO- of a inagnetized screw driver, touch the
S C O P E S - E X C E S S I V E D E L A Y BE- base of the crt near \vliere it joins the
FORE CRT BEAM COMES O N glass neck. \Vliile moving the tip of tlie
Time-delay relays used in tlie above os- screw driver around tlie available circum-
cilloscopes delay their operation for ap- ference of the crt base, check for dark
proximately 45 seconds after tlie power areas within the tiefocused spot on the crt
switch is turned on. This brief delay al- face. If dark areas are observed the crt
lows the tubes to warm up to near tlieir is suffering from low emission.
operating temperature before the dc oper- If either tlie 5642's or the crt are low
ating voltages are applied. At the end of in emission they should be replaced.
this delay period the cathode-ray beam
should appear on the face of the crt. T Y P E CA P L U G - I N U N I T - LACK
A more lengthy delay (two or more O F DUAL-TRACE DURING WARM
minutes - or up to 30 minutes in aggravat- UP
ed cases) can very often be traced to low Type CA Hug-In Units, serial numbers
emission by one or both of the 5642 tubes 101 through 34790, may exhibit a lack of
Figure 3. Schematic of K601 relay after per-
forming Delayed Pulse Modification. in the crt gritl supply and the crt liigli- dual trace during tlie period when the in-
voltage cathode supply. O r , it may be due strument is warming up. Tlie problem is
T Y P E 575 T R A N S I S T O R C U R V E caused by V3382. This 6AL5 tube in the
to low emission in the crt itself.
TRACER-VIEWING FIELD EFFECT switching circuitry has its cathodes re-
T o determine if the 5612 tubes are at
TRANSISTORS' CURVES turned to the -150 volt supply through
fault, remove the g r o m d strap from the
Nortnally, a Type 575 Transistor Curve crt-cathode connector located on the rear n 1.8 meg resistor in the oscilloscope via
T r a c e r is limited in displaying a family panel of the oscilloscope. Patch a cord pin 16 of the interconnecting plug. Tlie
of curves for an F E T (field effect tran- fro111 the calil)rator output to tlie crt-cathode 1 8 meg resistor provides a current source
sistor). \Vhen the S T E P S E L E C T O R connector and feet1 in 10 volts of calibrator for the 6AL5 that tends to balance the
control of the Type 575's Base Step Gem signal \\Tit11 the sweep free running you multivil~rator plates (V3375) in the CA
erator is set to the maximum (200ma) should now see a motlulated trace on tlie unit; both halves saturate and prevent mul-
position it will not completely cutoff the face of the crt. Advmce the calibrator tivi1)rator action.
FET. control tlirough the 20, 50, and 100 volts A 330 k, T/l w, 1096, compositioti re-
A simple modification is to place a 10 k, positions. If the modulated trace remains sistor (Tektronix part ~iumber 316-334)
G w , 1% precision resistor between tlie 011 tlie crt face the 5642 tubes are most added between pin 5 of V3382 and f225
base and emitter terminals of tlie Type 575 probably functioning properly. volts will cure the problem.

T Y P E 111 P R E T R I G G E R P U L S E G E N - T Y P E 4S2 D U A L - T R A C E S A M P L I N G This tnodification applies to Type 4S2
ERATORS-PULSE WAVEFORM IM- U N I T S - T R A N S I E N T R E S P O N S E IM- instruments with serial numbers below 301.
PROVEMENTS PROVEMENTS Order througli your local Telitronix Field
This modification improves tlic transient Engineer, Field Representative, Field O f -
hi^ modification reduces overs~loot,riIlg- fice or Distributor. Specify Telitronix
ing, and other aberrations in pulse response and retlil~eS ringing 011 fast-rise
part number 040-379.
waveform. I t also improves the risetime s i ~ l l a l s the Type
of the negative ~ u l s e . 1. Replacing Gate (bridge) diodes with T Y P E 53/54C A N D T Y P E C.4 DUAL-
Pritnarily, the modification consists of closer-matched and lower-capacitance T R A C E P L U G - I N UNITS-SLAVE TO
replacing the Avalanche transistor (Q84) diodes. AUTOMATIC DISPLAY SWITCHING
and reworking the associated circuitry on 7 Tliis modification allows Cliannels A and
-. Making the sampling bridge compensa-
the etched circuit board. New "transition B of either Type 53/54C, serial numbers
tion networks adjustable.
pieces" are used to connect tlie Charge 3710-1111, or Type CA, serial numbers 101
Line arid Output Polarity coaxial cables 3. Substituting 200 $2 resistors for tlie through 64009, to be slaved to the respec-
to tlie hoard. ferrite bends hetween srunpling I~ritlge tive sweeps of the Type 547 Oscilloscope,
Parts Replacement Kit 050-216 is also and Nuvistor gritl. wlien the Type 547 is operated in A A L T
included to replace the O U T P U T P O - 4. Terminating the strobe pulse lines with R mode. Tlie modification does not change
L A R I T Y switch and Charge Line cable. 100 9 resistors. the operation of tlie Type 53/54C or Type
This modification applies to Type 111 CA wlien operated in any other instru-
instruments with serial numbers below 800. 5. Adding grid-bias balancing potcntiom- ment.
Order through your Telitronix Field En- eters for each Nuvistor.
Ostler tlirough your local Tcktronix Field
gineer, Field Represetit:ltive, Field Office 6. Decoupling the -100 and +300 volt- Engineer, Field Iicpresentntive, Field Of-
or Distributor. Specify Tclitronix P a r t ages to tlie Sampler and Gate-Gcnera- fice or Distributor. Specify Telctronix part
Nu~nber040-392. tor circuits. number 010-391.
by Russ Myer
Tektronix Advertising Dept.
T h i s article /orr?z.s a couccptlta/ hasis /Or the ~ i n d c r s t a ~ z n ' i ~ l y Spec-
t r u w i l ) ~ a I j s i s ,tlz1i.s prcparil~g tlze rcadcr / o r flzc scucral adva~lzcd C ~ L ~ ~ Y K S
available of1 flzc subject zurittr)z or1 tl7c E~lgirlccriuylevel

Part I These radio transmissions take various tlie signal might look like f i g r e l a .
electronic configurations, ranging from sin-
gle-frequency carriers to coniplcs signals
produced i)y changing these carriers in am-
plitude, f requcncy and phase.
.At any given motnent, there is an incretl-
ihle amount of activit) ~ i t l i i nthat portion
Iiegartlless of tlie shape of these signals
of tlie Electrom:ignetic Spectrum tl~;it \ \ c
:mtl how they \vere protiucetl, o r "motlu-
call tlie 1i;idio Frequency liantli Tlieie
I;~tetl", e:tcli one c:ul I)c sel);~r:~tetl intlivi-
hinds rruigc in freclucncy from :il)out 1.5 kc
tlu:il sine ivavcs. E : ~ l isine w i v e represents
to 750,000 M c .
:I single irequency. '' cx:uiiine the co~ii-
?Issu~neyou have :I special radio receiver position :\titl quality o i ;I signal, you \vould Figure l a . Conventional oscilloscope display of
cap:hle o i tuning over this entire r:inge. :It iintl it very Iielpiul to extract e:\cl~ indivi- 1000 kc carrier modulated by o 1000 cps tone.
llie lower etitl. you'll find m;rritimc ship-to- dual sine \v:\vc tli:\t it contailis :ind display
shore, a i r c r a f t point-to-point, higli-po\rcretl it nlonc 011 ;ui oscilloscope. Seeing all the
government :ind commerci:il lransoce:inic sine waves in :i "group" picture, e;icli stantl-
signds. Tuning Iiiglier i l l frequency, ~vitliin ing nlo~ie,\vor~ltl euxl~lcyou to ;~n;ilyzethe
tlie i;\mili:ir iM-to-1600 1;c 1,ro:idc;ist Inntl, complex sigtial. T l ~ cinstrument tli:~t per-
tlozens of commerci;il r:ttlio stations compete forms this t:isli i o r you is called a Spectru~ii
for your attention. ribovc these, you'll fitid :\n:rlyzer.
more slii1)-to-sliore, :ind, c o n f i ~ ~ e tto re1:t-
l ,. l o use :\n ex;tml)le of :I familiar but
lively sm:ill portio~isof the spectrum, tliou-
s:inds of "1i:ini" r:itlio oper:itors pursue tlleir complex I\-:iveform \vhicli could Ile reduced
electronic entlenvors. Also, i n t e r s p : i c e t l to itidi\;idu:~l sine waves f o r :uialysis, con-
tlirouglior~t tliis short-n.:ive hand, you \rill sider :in A M radio st;\tioii. .\ broadcast
hear mucli :iir-ground :\ctivity, government t ~ t n s m i t t e r r:itli:itcs :I single c;irrier i r e -
poin-to-point, ni:\ny foseign I~ro:itlc:tst st:i- cluency from its :intcnn:i. Inlelligencc
tions, the i70ice o i 11meric:~ (:mi Mos- (slxecli, music, tones, etc.) is superiniposed
c o w ! ) , police r ; d o hroadcnst stations, ant1 o n this carrier, varying its :uiiplitutle at :in
some expc~-imental~vorli. ;audio rate. .\ssume tlie station is transmit-
ting ;\ 1000-cycle test tone. T h e cnrrier fre-
Still Iiiglier in frequency, you'll iintl tcle- cluency of the st:ttion is 1 Mc. This carrier Figure lb. Display of same signal using Spec-
vision st:ttions, st:\rting ;it .54Mc, F M st:i- is comhined in the iiwl st:lgc o i tlic tr:ms- trum Analyzer.
tions ;iimve 88 Mc :end more television :il)ove mitter n.it11 the 1000-cj.clc to~ic. Tlie :Inten-
174 Mc. Tile nre:i nlmve 400 M e , once con- x t , lioircver, through tlic process of "modu- Elcctric:\lly, the carrier is still occupying
sitlered experi~iient:~l, produces tiiyri:td six- I:itionl', is hro;itlcasting not two, hut f l r r r ~ ~ the I-Mc spot in the spcctruiii. I sign;ils. iTie\ved on n convcntio~i:tl scope. 1000 cycles belo\v tliis irequency, however,
at 999 kc, you will find a new signal, called
the "lower sideband". 1000 cycles above the
1 - M c spot, at 1,001 kc, you'll find another
signal, identical to the one at 999 kc, called
the "upper sideband". Tlie separation is ex-
a c t l y equal to the modulating frequency -
the 1000 cycle tone. The Spectrum Analyzer
is capable of displaying these three fre-
quencies, individually, on the screen of a
cathode-ray tube. Tlius, the component f re-
quencies may be individually studied, or
"analyzed". Figure l b shows how the Spec-
t r u m Analyzer would display them.
There is nothing

+' h *
~ 9ii,i* "4 a
( difficult about the
overall operation of
t i e analyzer, Tfie sig-
nals which we will
use as examples, how-
ever, must be fol-
lowed in detail
through the d ~ferent
sections shown in the block diagram T o
understand tlie conversion of input signals
to signals of lower frequencies, you will
f i n d it helpful to perform the simple aritli- Block diagram of typical Tektronix Plug-In Spectrum Analyzer.
metical computations dealing with the mixer the T ) p e L 2 0 , will analyze frequencies
frequency that was equal to the mathemati-
and i f (intermediate frequency) sections from 10 Mc to 4 G c , in 5 bantls. Ii'e will
cal difference of the input signals. Also,
There are several ways that a signal they found a 4th frequency in the output - consider the range of frequencies coverctl
c a n be broken down into component sine one equal to the sum of tlie two original by band 2 of the Type L-20, roughly 230
waves. One method is to i~itroduce the signals. They applied this pri~&ple to the Mc to 900 Mc.
signal to a stack of filters, the inputs of superheterodyne receiver, like one you prob-
which are paralleled. Each filter is tuned, :iblp have in your home today. Tlie follow- Refer to the block
in successsion, to a slightly different fre- ing example illustrates this concept, so diagram of the anal-
quency than the others. The output of each necessary to the understanding of Spectrum Incotn~ng sig-
filter will contain only tliat portion of the Analyzers. are ~ n t r d u c e d
input which corr