DOCTORAL DISSERTATION SERIES ■m siltlLIWlOH 6P blfLtCTMNl f£6M nm m sm nut tim w to p m ptsme eu sm cu pvlsis o f 0,66b M /e& 6S eC 6m MkATtOH AT M P P L P m m PATES autho r F&ANft MOPAAT PELTON UNIVERSITY DEGREE H MICfl' STATS COLL I DATE PUBLICATION NO. m 3 (e>70 m u UNIVERSITY MICROFILMS A N N ARBOR /fS i ■ MICHIGAN UTILIZATION OF REFLECTIO NS FROM TRANSMISSION LINE ELEMENTS TO PRODUCE POSITIVE ELECTRICAL PUL SES OF 0.006 MICROSEC­ ONDS DURATION AT HIGH REPETITION RATES By F r a n k Murray P e lton A THESIS S u b m i t t e d to the School of Graduate Studies of Michigan State College of Agriculture and A p p l i e d Science In partial fulfillment of the requirments for the degree of DOCTOR OF PHILOSOPHY Department of Electrical Engineering 1950 ABSTRACT The art of short pulse production has previously been limited by the lack of a suitable Interstage Inversion element usable for pulses of less than 0,05 microseconds duration. This paper treats the use of the short circuit terminated transmission line to accomplish this inversion. Inherent In the process when applied to triangular pulses Is a desirable reduction in pulse duration by a factor of two. Thus it is relatively simple to accomplish an overall reduction of pulse duration of more than two to one per amplifier stage. Application of this method was made in a pulse generator and amplifier capaKT e of oroducing 100 watt positive triangular pulses of 0.005 microseconds duration at a pulee repetition rate of 800 kc. The study of the production of pulses les«* than 0.005 microseconds and at repetition rates of greater than 800 k c . Is contemplated as the study made shows no limitation at these values. ill PREFACE The development of this thesis Is the outgrowth of the need for a very short gating pulse in the production of a special waveform similar to a recur­ rent sawtooth voltage. The inversion method with the shorted transmission line as applied to short pulse generation is an original contribution to the art by the author. The originality is in evidence by the lack of previous work in the literature and also by the anxiety expressed by the U.S. Armed Forces in obtaining the method. The reader should be cautioned that the inversion principle as applied to pulse production is being submitted to the Office of Naval Research for Patent Application in the author's name, and any commercial utilization other than continued research will be prohibited in the event the patent is granted. This method has been made available to the Office of Naval Research and through them it will be made available to the U.S. Armed Forces for National Defense uses, royalty free. The author wishes to thank Dr. J. A. Strelzoff for his help in editing this thesis and to both him and Mr. Noah Kramer for their suggestions as to method of mathematical Spenoe solution, and to Dr. R. D. for discussion relative to the interpretation of certain technical phenomena. TABLE OF CONTENTS A b s t r a c t .................................... Ill P r e f a c e ...................................... iv Table of C o n t e n t s .......................... v CHAPTER I page 1.1 Introduc tion .................................. 1 1.2 Peak Amplif iers ............................... 3 1.3 "Normally on" Inversion S t a g e ............... 4 1.4 Meed for a New Inversion M e t h o d ............ 5 CHAPTER II 2.1 The Characteristics of the Short Circuit Terminated Transmission L i n e ............ 7 2.2 Analysis for Square Wave Applied Voltage... 9 2.3 Analysis for Applied Voltage of Triangular P u l s e ....................................... lO 2.4 Differentiation by a Shorted L i n e ........... 10 CHAPTER III 3.1 experimental Results for Square Wave Applied V o l t a g e ............................ 14 3.2 Differentiation Action of Shorted L i n e .... 14 CHAPTER IV 4.1 The Typical Pea k Amplifier Stage............ 19 4.2 equivalent Circuit for the Peak A m p l i f l e r ..21 4.3 Method Solution U s e d ......................... 24 v page 4.4 Leading kdge of the Negative P u l s e ........ 24 4.5 Trailing kdge of the Negative Pulse* ..... 25 4.6 Output Impedance of the Peak Ampllfier....28 4.7 Solution with Shorted Line Attached....... 29 CHAPTER V 5.1 Photographic Data on Reflection Process*..32 5.2 Delay Line D a t a ..............................35 CHAPTER VI 6.1 The pulse Generator U n i t ....................41 6 . 2 Data on Pictures of the U n i t ................ 53 CHAPTER VII 7.1 Conclusions........... 63 7.2 Suggestions for Future S t u d y ............... 64 Appendix A — Short Pulse Measurement.......... 66 R e f e r e n c e s ........................................ 69 Bibliography,................ .................... CHAPTER I 1.1 Introduction The production of very short pulses Is Important in many fields at the present time. The art Is in a stage of rapid advancement due to the recent need for pulses in the range below 0.01 microsecond duration and at repetition rates greater than 1 Mc/s. The methods previously used are un­ satisfactory and most have serious 11mllotions. The author has developed a method utilizing a short circuit terminated transmission line as the inter­ stage inversion element in a typical pulse peak amplifier. The previous limits In short pulse genera­ tion have been considerably surpassed, and 100 watt positive pulses of 0.006 microseconds duration have been produced at an 800 kc. repetition rate by a pulse generator incorporating this new inversion process. A brief account of existing methods will now be given. At the end of the war, the Radiation Laboratories published a series of books listing most of the work done during the war period, 1940— 45. U s ing tnese as a starting point, the common pulse generation methods will be considered. They may be generally classified into multivibrator, blocking and regenerative, delay line amplifiers and line type pulsers, non-linear p e a k e r s , and peak amplifiers. Chance"*" mentions the production of pulses °f 0.1 microsecond by certain multivibrator circuits and shows some w h i c h are suitable for 0.5 microseconds. These are limited by the pulse power available and the presence of both tube and stray capacitances. The loading affects the transient operation and thereby causes an Increase In the pulse rise time and the overall pulse duration. Blocking oscillators and regenerative oscillators are limited primarily by the characteris­ tics of the pulse transformer which is used. Pulses ae short as 0,1?. microseconds duration were produced by cascading two such oscillators. was .03 for 200 v. pulse. The rise time At the present time, the R a y t h e o n transformer type UX 7307 is advertised as capable of the production of pulses of 0.05 micro­ seconds duration, and is generally considered to be about the limit for blocking oscillators at this time. Delay line amplifiers or tall clipping circuits in general, require pulses with fast rise times on which to operate, and in themselves do not increase the rise time of the pulses. This limits the magnitude available before the clipping action begins. Line pulsers are capable of producing pulses of 0.01 microsecond but are limited in their application because of the slow repetition rates n e c e s s i t a t e d by the deionization time of the thyratron tube used to initiate the p u l s e . 3 recently Non— linear p e a king colls have been used 4 to produce pulses of 0.01 microsecond dura­ tion but the p o w e r output is limited by the very small di m e nsions necessary to obtain the short response. P ea k amplifiers also have been used to pr o duce pulses in this range. Negative pulses of 0 .02 microseconds duration w i t h a rise time of only 0 .003 microseconds have been produced w i t h a sine wave pea k ampl ifier.5 M e n t i o n has been made of the use of this type generator up to 5 Mo. repetition 0 rates. If a suitable method of pulse inversion was available, p e a k amplifiers could produce pulses which inco r p o r a t e d the full gain bandwidth possibilities of a given tube. 1.2 P eak Amplifiers Pe a k amplifiers are ideally suited to p u lse amplification and pulse shortening. They c o n s i s t primarily of a grid biased amplifier tube w i t h a suitable plate load impedance. V i deo compen­ sated plate circuits may be necessary in some cases. A positive grid wave form, usually a triangular pulse, or a sine wave is biased so that only a desired form, p o r t i o n of its positive peak causes conduction in the tube. The selection of a tube with very high mutual transconductance, such as the 6AG-7, together with high dissipation provide an efficient switch action at quite low impedance when operated in the positive grid region. The negative pulse appearing at the plate is then inverted by a suitable means and the process is repeated. rhus the total gain-band width factor of the tube used is utilized. For short pulses at low duty rates, the applied plate voltage and the Instantaneous power in the pulse may be many times the continuous rated values for the tube. Short negative pulses are very useful as markers and time modulators in cathode ray tube dis­ play work as their apparent polarity may be controlled by proper connection to the circuits involved. How­ ever, for most applications such as triggering, further shortening, gating and pulse amplifier circuits in general, a positive pulse is required. Thus a suitable means of inversion of the negative plate pulse is required. Assuming that pulses are desired which would not be passed by the available pulse transformers generally used for this process, other methods of inversion will be considered. 1.3 The "Normally On" Inversion Stage A "normally on" amplifier stage, that is one whose tube is normally in a conducting state, when driven with a negative pulse produces a positive plate output pulse equal to the power dissipated in — 4—• its normal state. This assumes the grid signal is sufficient to drive the tube to out-off bias. The disadvantage of this inversion process is immediately apparent in the d.c. plate supply power loss in maintaining the tube at high dissipation. If this process is to be used, mention should be made that if the negative peak of the input signal is "taken off" with a non-linear diode before application to the amplifier tube, considerable shortening is ob­ tained because of the diode's characteristics. Al­ though no experimental evidence has been obtained, it would appear feasible to cascade such diode stages, each with its own bias, to cause further shortening of the negative pulse. This would entail having a large negative pulse with which to start. For most applications, the inefficiency of this method prohibits its use. 1 .4 The Need for a New Inversion Method Other than the above methods, no standard method of pulse inversion is available. Pulse trans­ formers will pass only about a 0.05 microseconds pulse and "normally on" stages are so inefficient that they cannot be used. It is apparent that if shorter pulses are to be produced by peak amplifie rs, some satisfactory method of inversion is necessary. To obtain a short time constant, i.e. the response of the tube and its associated inversion network to — 5— to produce the d e s i r e d pulse (volts/sec p e r volt applied) , the prod uct of the impedance and the shunt capacitance must be made low. Thus both of these factors are made as low as possible as is consistent w i t h reasonable design considerations. The remainder of this thesis will consider the use of the short circuit terminated line to accomplish this inversion. — 6— transmission C H A PTER II 2.1 The Characteristics of the Short Circuit Terminated Transmission Line_________ Consider the ideal transmission system7 shown in figure 1. The voltage v(t) is assumed to be produced by an lmpedanceless generator. It has been shown® that the voltage at any p o int along the transmission line is = — 7T. + 1 Now if either Zg or + -' is less • ] than the characteristic impedance of the transmission line, a negative reflection will result. voltage pulse v(t) tion arrives, Furthermore, has terminated before the reflec­ the voltage e(o,t) will show polarity opposite to that of v ( t ) . As this is the desired result for this application, phenomena. if the impressed use is made of this For maximum inversion , the output end of the line is short-circuited, i.e. Z^.“ 0, and Zg is made equal to the characteristic Impedance of the transmission line to suppress further reflections. The voltage e (o,t) then becomes M >- M o fsl 2.2 Analysis for a Square Wave Applied Voltage For an elementary analysis, assume that the applied voltage function is of the form e* o t^o ^ - | o <. t. e= o f< * ^ C o n s idering the delay time 2d in equation (2) , three cases become of interest: ( 1) For 2 d > 1 , the reflected pulse appears at the input of the line after a delay two pulses may be observed quite i 8 shown in Fig. (2) For 2 d < 1, time of 2d— 1 and the independently as 2. the reflected pulse appears at the input of the line while the voltage v( t) is still applied and the addition of these produces a voltage e(o.t) such that ^ . e= o ± < o e= i o < £ < z*A e =o zd< * < / e= -i i < * < *4 e= o v-d < £■ Thle is shown in Fig. (3) If 2 d » l , the voltage rises instantaneously from 1 to -1 at t«l as is shown in Fig. obviously provides 3. the greatest voltage 4. This change per u n i t of time for any of the three cases bei ng con­ sidered. 2.3 Analysis for Applied Voltage of Triangular Pulse When a practical application of this system is attempted, the transient response of the networks enters into the consideration and this maximum applied voltage change is important in that it doubles the rise time of the resultant pulse. This effect may be more readily seen if the addition of 2 triangular pulses is considered. in Fig. 5. This is shown In this interpretation, it should be pointed out that the leading edge of the triangular pulse is produced by the charging of the shunt capacitance of the associated networks during the time an effective square wave voltage Is applied. The trailing edge is formed by the discharge of the same capacitance after the effective square wave applied voltage is terminated. Thus this Is not a contradiction to the cases discussed In Sect. 2.2. 2.4 Differentiation by a Shorted Line If a transmission line is very short compared to the Input voltage frequency components, the shorted line appears as a capacitance. The input termination together with the reflected wave, which becomes apparent as the charge impressed on the equivalent capacitance, forms a differentiating circuit similar to the familiar R —C differentiator. 10- Z P>1 h 2P1 k 20*1 J **1 I I I I I I I I FI G. 2 FIG. 3 FIG. 20* I FI G. 5 FIG. 6 CHAPTER III 3.1 Experimental Results for Square Wave Applied Voltage_______________ In equation (2) reflection from the sending end was suppressed by the Input termination of ZQ , the characteristic Impedance of the transmission line. In the practical application, however, it Is impossible to obtain an impedanceless voltage source and hard to obtain a source with an Impedanceless than the ZQ of a natural line. Also the reactive component of the Internal Impedance of most physical generators is large enough to show considerable deviation from the Ideal case. To partially overcome these difficulties, an artificial line was constructed with characteristic impedance of 1000 ohms, and driven by a pulse genera­ tor as shown in Fig. 7. The three cases considered in beet. 2.2 are shown in the oscillographic photographs in Figs. 8, 9, and 10. The attenuation of the line accounts for the decreased amplitude of the reflected pulse. 3.2 Differentiation Action of bhorted Line The differentiating properties of a short circuit terminated electrically short transmission line were also experimentally demonstrated by the setup in Fig. 7. The applied voltage and the -14- L o & z PULS t w t innnj— innnr c I GEN. ARTIFICIAL FIC. line r^Tnrzr6 voltage e(o,t) are shown In the oscillographic photographs In Fig. 11 . -18- FIG. 8 CHAPTER IV 4.1 The Typical Peak Amplifier Stage As mentioned in Sect. 1.2, the choice of a tube for peak amplifier application is primarily determined by the transconductance (Gm ) and the plate and grid dissipation. In addition, for very short pulse work, a large gain-bandwidth product is desired. The 6AG7 tube, which is quite well suited for this application, was used for the experimental work in connection with this problem. 0.05 microseconds duration, For pulses less than two 6AG-7 tubes were paralleled to Increase the available dissipation, and the resulting circuit is shown in Fig. 12. components used are listed below: R1 2k °2 0.006 rafd. *2 100 to o 0.006 mfd. R3 lOk C4 0.006 mfd. R4 50k C5 0.05 mfd. V 1 & v2 6AG7 Plate supply voltage « 400 volts Screen supply voltage *= 250 volts This circuit was driven by a IOC volt positive triangular pulse at an impedance of 100 ohms. The Ep E Sv OUTPUT B HI -150 V FIG. 12 I B 4.2 Equivalent Circuit for the Typical Peak A m p l ifier ___________________ The procedure normally used in deriving an equivalent A . C . circuit of an amplifier containing a pe n tode tube is to replace the tube by a constant current source In eerles with the equivalent plate resistance of the tube. however, For positive pulsed tubes it is more convenient to replace the tube by an equivalent plate resistance in series with a switch, leaving the plate p o w e r supply connections intact. This is Justified as follows: Reference to the positive pulsed grid Q o p e r a t i n g characteristics for the 6AG7 tube. Figures 13 and 14, show that its plate resistance is nearly constant after the grid is 25 volts positive. a m p l i f i e r in Figure As the was driven by a 100 volt positive triangular p u l s e , only about 1/8 of the period of the d r i v i n g pulse is used to produce the leading edge of the effective square plate resistance function. that time period is neglected, If the tube may be re­ p l a c e d by an equivalent plate resistance in series w i t h a switch. per Determining the average plate resistance tube from Figure 14, of the paralleled stage, circuit. this value is halved because and used in the equivalent The equivalent circuit obtained w i t h these assumptions is shown in Figure 15. capacita nce of the paralleled 6AG7 stage. tate calculations, the circuit of Fig. on the nodal basis in Fig. 16. is the To output facili­ 15 is represented • ■ ■ Epp IN V ............. O t....... Fig . i3 - . U I- T - S e i N V & 1TS 1........ r.......•• f....... B e E FIG. 15 €>- 9 0 FIG. 16 ■0- 4,3 M e t h o d of Solution Used Using the method of LaPlace transformation3 the transform of the output of the peak amplifier will be found first without the line connected. Next, back the transform of the impedance function looking from the output terminals is calculated. i'he app l i c a t i o n of Thevenin'a Theorem to this system allows an analogous solution to that used in Sect. 2.2: it is considerably complicated by the equivalent input impedance Zg. After the comolete solution has been found in terms of tne LaPlace transform, inverse the transformation will be calculated obtaining the resultant pulse waveform as a function of time. 4.4 L e a ding kdge of Negative Pulse The nodal equations for Figure 16 are X = Lg,+S,+ Cc.+ c,') ,1 v, It) - Cj , Witt) (4.1) (4.2) 0 = - CjT v . t t ) ( C j * c3)p]Vitrt where p«d/dt Taking the LaPlace transform of equations (4.1 & 4.2) - - * ^C‘+ - £i+c*}e, - C24V,C0+ Cget ( 4 *3 ) ( 4.4) Solving (4.3) Vt vs) XL and (4.4) for s) ^ » S +• (4 .5 ) A i't Bs+ c where + Cc,+ cty c , t c5) -25- (4.9) (> VVvVvVV ei o o VWsAvV" o o o where A - cje,+ CCl+c ^ C . f Ca)e,- 2 e ^ J e = c , ^ C C , + 0 + (.<,* $,\Ct+ c»+ i- q^x D > C,* vCCe+t.Xc.e-O E. • (c,-v c*^ +• (.<*•»•5iXc*+ q ) F = C , (G,+0 Taking the Inverse LaPlace transformation of v.lt^ If _ fr-y.Xd-T.) £'7,*_ J r.t, (4.9 )12 (4 .1 0 ) (^-<3 0 W - £) r.is.-y.'i cf, t Y , - < 0 At time t=2*t , (2t*duratlon of the input driving pulse) the equivalent circuit has changed to that shown in Fig. \7 • v^( t) and v£>(t) have initial values e^ and eg calculated from equations (4.7 and 4.10). The equations for F i g . 17 are T - £( (c.+^V) f J (,-tr) - C2 p V, Ct) O = - c2 p V, (*) pi -J-+ ( <*1 + £3 }p] VaHr) (4 .11) (4 .1 2 ) The LaPlace transformations of the above are — = 1 ^,.+cc,+ c.>s3 w,(s) .(c,*y«'r Cjsvjto * ca< S (4 0= -C&V'{3)+ CiC,'+ t ^ > . ^ ^ ^ c3>)slv3is)-(C2+C3>)ej .13) (4 .14) whereby va'cS ) - ± ^ i u L i= A 1s*~+ Bs +-C ^ L ±-r > * % (s+*-')( s + (3')J -27- i < * .15) where a;- B -t Cc.+C») c ’= c7,$2 The Inverse transformation yields V - t ) = g i r tf;-*') t-*'t+ (jilf? £'p t (4.15) A Lg>'-*^ -1 This is the trailing edge of the pulse produced by the amplifier circuit. 4.5 Output Impedance of F i g . t€> The impedance if the network looking toward the tube from the input terminals of the shorted line is \ £ = ( 4.17) + A, s where v- vC| + = Cq + / i) C| 4 Ci C, z Cl■*" -28 4.7 Solution with Shorted Line Attached Applying Thevenln*a Theorem to the system, the equivalent voltage Is that derived as the output pulse from the tube circuit without the line attached, and the equivalent series impedance is that in (4.17) with the change 0^ - t < 2 't' G, = O thus shown -tftX vg If used with Z, and Vg Is used with Z* From Sect. 4.3, (4.6) s*"+ S s + C, The solution of Fig. E * F a, 1 is shown by G-oldman C ~ ra©-eT 1 U ^ - to be Svr(a«l-») l >) « *« (4.18) LH.+ i T j l The Insertion of the terminal conditions in this caae give the voltage across the impedance Z E (»V*)* (a s ' + B s + c )( f O to be _ ^su-A 4 • + U i * C’ s+ A '* ( sfi, * + (4.19) * 6 +’ I ^1-+ C3S * -29- where the subscript 1 has been added to distinguish the coefficient in the expression for the impedance. Further expansion gives = 4 I -’ A + cj9 ( M p O + fl/s+w.l +^(s-*p.^h+*<)(s-tp) -2*ird t -25u vr i >3- i >V >i «£i X i/ Vd » V*“* •SfO* i <5/ J S C' « <3 Lf} it I i I >o 3 £ ^puatlon 'v>3 ( 4 . C l) VT& i uo CM it it ii *|K _x> JCk I i :* r f -p 5 1 -31— CHAPTER V 5.1 Photographic Data on Reflection Process The fastest sweep available In the labora­ tory was 0.1 microsecond/cm. data, and to provide Interpreted the pictures were taken using a 0.1 microseoond negative triangular pulse. Figures 21, square negative pulse 22, and 23 show an approximately of 0.12 microsecond duration, the reflection from an open line of 0.20 microseconds delay and the reflection from the same line when shorted. The reduction in amplitude is due primarily to the mismatch at the termination and not to the attenuation in the line. This is obvious on noticing trie second reflection starting at O.-iO microseconds; in both cases it is positive as it should be when evaluating the reflections on the basis of the diagram and the boundary reflection analysis^4 shown in Fig. 20. It should be noted that for many cases, this type analysis is sufficient. Readjustment of the pulse generator produced the pulse shown in Fib. 24. This is approximately triangular in shape and of duration of about 0.15 microseconds. edge, i.e. It should be noted that the leading that part of the pulse produced by the effective square negative pulse impressed on the network, takes about 0.05 microseconds to come to maximum amplitude. The series of pictures In Figures — 32— -E OPEN 2 )»• J - * + e /4 'i / 4 - E/ 2 ^ + E/4 Zo -E 2 >► c/ *2 SHORTED ►» — - -- E/2 + E/4 1 + E/2 : -E/4 >► - ^ -FE/4 FIG. 2 0 25 through. 30 show the effect of the variation of the length of the delay line on the resultant pulse waveform. P a r t i c u l a r notice should be given to the time at which the reflected pulse produces in polarity. in Sect. the change These all co rrespond to case 2 discussed 2.2 and show definitely the cancellation p e r i o d and the change in pulse length predicted by this simple addition of the original and reflected pulses. The entire series of pictures in Figures 21 through 30 were taken w i t h vertical calibration at 240 volts/div. and the horizontal calibration at 0.2 microseconds div. A composite multi-exposure is shown in Fig. 31. The wave forms are distorted due to the a m p l i f i e r in the oscillograph but the relative amplitudes and time displacements are sufficiently accurate to show the variations expected. Tabular data of line lengths is shown on the following page. 5.2 Delay Line Data The following data gives the length of the delay lines used to produce the oscillograms in Figures 25 through 30. Figure Length 25 10.5 ft. Calc. Delay 0.0315 usee 26 6.5 0.0195 27 3.25 0.00974 28 2.08 0.00623 29 1.38 0.00413 30 CO • o 0.00252 The delay was calculated on the basis of two way transit time and a velocity of propagation factor of 0.695. The lines used in Figures 26 through 30 are the lines used in conjunction with the pulse amplifier unit described in Chapter VI. 35- CHAPTuiR VI 6.1 The Pulee Generator Unit The principle of pulse inversion by shorted transmission lines has been applied in the construc­ tion of a generator capable of producing positive pulses of 0.006 microseconds duration at repetition rates of 400 k c . to 800 k c . The pulse generator and all of its associated parte were designed and con­ structed by the author. Although considerable time was spent in this work, the inversion method and the end result of its application is of primary Interest in this thesis, and therefore the details of design and construction have been omitted. The pictures and diagrams are mostly self explanatory, but brief explanations have been Included where necessary. The basic pulse was obtained by applying suitable wave shaping circuits to a sinusoidal waveform. rhls choice was made primarily because of the necessity for a small amount of Jitter in the observation of very short pulses. Two stages of peak pulse amplification utilizing pulse trans­ formers as Interstage coupling provided a negative 100 watt triangular pulse of approximately 0.1 micro­ seconds duration. The pulse repetition rate was variable from 400 k c . to 800 kc. In addition, a cathode follower take off -41- stage triggered a multivibrator square wave generator operating at about 200 kc. This gated a linear saw tooth generator whose output was amplified and used as the sweep for a built in pulse monitoring oscillo­ scope. The square wave output of the multivibrator was also used as a blanking signal and applied to the grid of the type 3BP1A cathode ray tube. The negative pulse output of the basic g enerator was inverted by means of a shorted trans­ mission line and further peak amplification was applied. Four more applications of this process u s ine the basic amplifier circuit of Fig. 12 with appropriate lengths of RG7U coaxial line resulted in a 100 watt output pulse with a positive duration of approximately 0.006 microseconds. This pulse duration was not a limiting value but was considered ample to show the merit of this method of pulse inversion in that it far exceeds the published re­ sults at this time. The block diagram of the pulse generator is shown in Fig. 32 and the voltage waveforms relative to chassis of the points indicated by the circled numbers are shown in Figures 33 through 50. Figures 51 through 58 show the constructional details of the units comprising the pulse generator and the method of mounting. r SQ. WAVE SWEEP RECT. RECT mr SWEEP CONTR. [ CONTR.TCONTR. CONTR. V. R. POWER SUPPLY PULSE AMPLIFIER 6.2 Data on Pictures of Unite Figure 5 1 . Front view of completely assembled pulse generator unit. bottom) panels included are: (top to (1) basic pulse generator and monitor oscilloscope, d.c. me (2) peak pulse amplifier and 250 volt regulated supply, (3) 450 volt d.c. power supply including 3 separately regulated channels. Figure 5 2 . Rear view of completely assembled pulse generator unit. coaxial make Note the connection of the reflection cables on the pulse amplifier chassis. the unit more compact, To these cables were coiled and placed in the side of the rack cabinet. F igure 53. Top view of basic pulse generator chassis. The C.R.T. intensity, focus, astigmatism and position­ ing controls are grouped around the shielded G.R.r. The high voltage Fig ure 5 4 . chassis. supply is seen Just below the C.R.T. Bottom view of basic pulse generator Filament, high voltage, and bias voltage connections are cabled and run next to the chassis, ihe signal wiring is done with bus bar to maintain low capacitance. Figure 5 5 . Top view of pulse amplifier chassis. A 250 volt, 250 mllllampere regulated supply is mount ed at the top of the picture. The remainder of the unit is the repetition of the basic peak pulse amplifier of Figure 12. -53- Figure 5 6 * Bottom view of pulse amplifier chassis, showing the type of construction used. The mainten­ ance of very low capacitance wiring was essential in p r o d u c i n g the fast rise times. Figure 5 7 . rop view of 450 volt power supply chassis. Paralle led rectifiers were used to obtain low imped­ ance. Three separately regulated output channels supply 200— 450 volts at 65 milllamperes each. Figure 5 8 . Bottom view of p o w e r supply chassis showing layout and method of construction. — 54— I dd m e ti © ri i I $ Vr.; % C H A P T E R VII 7,1 C o n c l u s i o n s The r e sults o b t a i n e d by m e t h o d are significant. simple, The this inversion inversion process is e f f i cie nt and p r o v i d e s a pulse duration r e d u c t i o n by a f a c t o r of two if desired. No dif­ f i c u l t y was e n c o u n t e r e d In o b t a i n i n g the results d e s c r i b e d in C h a p t e r VI, and the author deems it likely that t h r o u g h ref i n e m e n t s in design this m e t h o d will enable less than 0.001 m i c r o s e c o n d and of re petition rates greater the p r o d u c t i o n of p u l s e s of durations than kiO Me. b e i n g relatively The factor of the inversion Ind e p e n d e n t of repetition frequency ( o t h e r than o v e r l a p p i n g of the d r i v i n g pulse) make it very useful lated pulse in an a m p l i f i e r for time modu­ systems such as P.T.M. and P.C.M. The general appl i c a t i o n s as those o b t ained by radar, vision, of p u l s e s such the uni t de s c r i b e d are numerous: time m e a s u r e m e n t , video will testing, co mputers, g a ti ng and Jamming, tele­ the use in the s t u d y of p h y sical p h e n o m e n a in other fields are but a few. Obviously , this a d v a n c e m e n t in the art of pulse production new and, provides field in w h i c h the o p e ning of an entire the lim i t a t i o n s of pulse duration or pulse r e p e t i t i o n frequencies have been removed to r e l a t i v e l y remote values. 53— 7.2 S u g g e s t i o n s for F u r t h e r S t udy It is the a u t h o r ’s Intention to continue this w o r k and to d e t ermine set up by physical tubes this method. the values of the limits F u r ther refinement of the c o n s t r u c t i o n and a p p l i c a t i o n of V.H.F. of low transit time should provide a considerable e x t e n s i o n of the values a t t a i n e d in C h a p t e r VI. It Is s u g g ested r e s e a r c h in this type In the u n i t described that others continue field in a d d i t i o n to s t u dying the a d a p t a t i o n of the m e t h o d d i s c u s s e d to the needs of short p u l s e s In o t h e r fields. may M any applic ations a r i s e w h i c h are not evide nt at the p r e s e n t Some suggest ions w h i c h m i ght be suitable (1) guides the study of ideas for theses m a t e r i a l follow: T r a n s m i s s i o n of short d.c. p u l s e s in wave (not radio (2) for time. frequency oulses) Radiation of short d.c. pulses from antennae. (3) Study of the p r o d u c t i o n of s h o r t e r pulses. (4) Study of pulse p r o d u c t i o n at h i g h e r repeti­ tion rates. (5) Design and d e v e l o p m e n t of m a r k e r generator of 0 . 0 0 5 m i c r o s e c o n d s d u r ation pulses at a 20 Me. r e p e t i t i o n rate. (6) (i.e. 0.005 mic r o s e c o n d s markers). D e sign of a variable w i d t h pulse generator at h i g h r e p e tition rates suitable for video of very w ide b a n d amplifiers. — 64— testing (7) D e s i g n a n d d e v e l o p m e n t of a h i g h speed synchroscope with calibration m e a s u r e m e n t of very (8) of Study o f timing m a r k e r s for the s h ort pulses* gas d i s c h a r g e u s i n g an a d a p tation the p u l s e m e t h o d d escribed. (9) seconds The p r o d u c t i o n of p u l s e s of 0 . 0 0 5 m i c r o ­ d u r a t i o n at v e r y h i g h powers. (Suitable for m a g n e t r o n keying) (10) Study o f n o i s e r e d u c t i o n in s a m p l i n g type r e c e i v e r s by a d e c r e a s e d open (11) Study of modulation (12) a colored tube using' these Study of tron and klystron time. television intensity short pulses. p u l s i n g c h a r a c t e r i s t i c s of m a g n e ­ oscillators for p u l s e s b e l o w 0.01 microsecond. (13) The d e s i g n and d e v e l o p m e n t of a device utili z i n g a coaxial in such a manner over tee, by p r o v i d i n g a c a n c e l l a t i o n to o b t a i n an i n c r e a s e d rise the m e t h o d d e s c r i b e d in (14) The d e s i g n this time thesis. a nd d e v e l o p m e n t of a h i g h r e solu­ t i o n r a d a r s y stem u s i n g the short pulses a v a i l a b l e by this method. An unlimitednumber physics and e l e c t r o n i c s may be applications will field of found and no d o u b t the vary widely. Communications Mich. of ideas in the State College to the author in care of regarding advancements and c o n t i n u e d r e s e a r c h in e i t h e r d e v e l o o m e n t or a p p l i c a t i o n will be a p p r e c i ated. -65- APPENDIX A Short Pulse Measurement The p r o b l e m of p u lse m e a s u r e m e n t Is one w h i c h r e q u i r e s a very s p e c i a l i z e d cathode ray l o g r a p h or synchroscope. For p u l s e s of* du ration below 0.05 microseconds, 20 inches/microsecond a sweep g e n e r a t o r a m p l i f i e r unit. This Is In Figures s h o w i n g the r e s u l t a n t p u l s e s Is of speed of at least Is deslreable. e v i d e n c e d by the p h o t o g r a p h s 50, oscil­ As 46 through from the oulse the base of the pulse the same o r d e r of m a g n i t u d e as the fluorescent line w i d t h on the cat hode ray m e n t s wer e tube, accurate m e a sure­ ob viously not possible. Sweep g e n e r a t i o n of the order of 20 to 100 Inches sweep p e r m i c r o s e c o n d requi re g e n e r a t o r s w h i c h must be operated by tubes. few This again limits P r o blems to a of Jitter very e v i d e n t and t r i g g e r i n g p r o blems must be handled in such a w a y the sweep that the observed p u lse t r i g g e r i n g pulse. (a train of s i mi lar p u lses be d e s l r e a b l e the ceeding thyratron the sweep recurrence t h o u s a n d c y c l e s per second. become that the use of line t y p e ^ to delay sweep w o u l d be is also For recurrent pulses evenly the sweep spaced) it would triggering pulse so initiated by the oulse pro­ the one b e i n g observed. This w o uld eliminate the n e c e s s i t y of u s i n g a delay line or d e l a y i n g -66- a m p l i f i e r w h i c h m i g h t introduce d i s t o r t i o n into the o b s e r v e d pulse. To the a u t h o r ’s knowledge, this f e a t u r e has not b e e n inc o r p o r a t e d in any of the c o m m e r c i a l l y a v a i l a b l e oscillographs. The new 5XP— series DuMont for pulse o b s e r v a t i o n has made the use of amplifiers unnecessary sensitivities for m o s t work. Def l e c t i o n s of 30 v o l t s / i n c h are a t t ain ed w it h direct connection T he tubes d e v e l o p e d by to the vertical de f l e c t i o n plates. overall a c c e l l e r a t i o n volta ge may be mad e as high as 20 Kv. This is the ideal system for fast pulse measurement. H i g h speed synchros copes are available from both DuMont and T e k t r o n i x , is p r o h i b i t i v e for general but in g e n e r a l , the price school use. Both of these s c o p e s c o n t a i n vertical d e f l e c t i o n amoliflers of the distributed constant to 75 Me. request. type w h i c h are useful DuMont can furnish Information on special T e k t r o n i x has a model ^517 in p r o d uction which sells Co^® p r o duces wave (line— type) for about 53500.00. The H e w l e t t - P a c k a r d b o t h d i s t r i b u t e d line type and travelll type am p l i f i e r s which are very useful for short p u l s e amplifica tion. C i r c u l a r sweeps radar "J" scopes^-^ could be u t i l i z e d for p u lse dura­ tion m e a s u r e m e n t s by central such as those u s e d in either by Intensi ty m o d u l a t i o n or e l e c trode d e f l e c t i o n tubes p r o v i d e d some m e t h o d of stabilizing* the c i r c u l a r sweep frequency (sine w a v e HF voltage) and s y n c h r o n i z i n g this source w i t h ■chat of the p u l s e r e p e t i t i o n tory tests showed d i f f iculty synchronization between 1 Me. b u t If the pulse frequency. In m a i n t a i n i n g stable two separate oscil lators at source (i.e. the sine wave d r i v i n g voltage In the pulse generator) used for the c i r c u l a r s™eep source, sible to m a i n tain The Labora­ c o uld also be it should be p o s ­ Jitter at a minimum. fastest c a l i brated sweep pr e s e n t l y a v a i l a b l e at M i c h i g a n State C o l lege is 0.1 microsecond/ cm. In the form of the T e k t ronix Model graph. M o s t of the pulse w o r k was done by terminating the m e a s u r i n g coaxial ray 511A oscillo­ cable d i r e c t l y at the cathode tube d e f l e c t i o n partes. The 5 C P 1 A , not p a r t i c u l a r l y adapted tube u s e d is a for this type use, so the results wer e not very good. of the r e f l e c t i o n s and oscillation s p r e s ent in the o b s e r v e d pulse are due to the leads Undoubtedly some from the termina­ tion of the coax and the d e f l e c t i o n plates. Those c o n t e m p l a t i n g the c o n s t r u c t i o n of a synchroscope this use are a d v i s e d to use for r e c u r r e n t pu ls e s transie nts. the 5J P — series or the 5XP— series If funds are available, w o u l d be most d e s i r a b l e for of tubes for single the 5 XP— series for e i t h e r case as the h i gher a c c e l l e r a t i o n v o l tage and d e f l e c t i o n sensitivity is necessary for very fast w r i t i n g rates and small d e f l e c ­ tion voltages w h i c h generally o c cur in short pulse o b s e r v a f ion. References 1. Chance, W a v e f o r m s , p. 181, M c G r a w Hill Boo k Co. 2. Chance, W a v e f o r m s , p. 236, M c G r a w Hill Book Co. 3. G l a s c o e , P u l s e G e n e r a t o r s , p. 4. L . W . H u 8 s e y , P r o c e e d i n g s of the Instltute of R a d i o en g i n e e r s , 38:40 5. Chance, W a v e f o r m s . p. 6. 7. 10. (Jan. , 1950) 351, M c G r a w Hill Book Co. W i r e l e s s e n g i n e e r i n g , 25:164 (M a y , 19 48) and 8. Goldman, electrical 9. 353, M c G r a w Hill Book Trans fo rmation Calcu lus and Transients, Wiley Chance, W a v e f o r m s , p. 214, G a r d n e r & Barnes, T r a n s i e n t s M c G r a w Hill Boo k Co. In L i n e a r Systems, W i l e y B o o k Co. 11. and 12. The Inverse L a P l a c e t ransfo rmation was o b t a i n e d by d i r e c t solutio n of the integral by the m e t h o d of residues. 13. Goldman, Trans format Ion C a l c u l u s & g.leotrioal T r a n s i e n t s , W i l e y B ook Co. 14. Schelkunof f, e l e c t r o m a g n e t i c W a v e s , D. Van N o s trand C o . , Inc. 15. boiler. C a t h o d e Ray Tube D i s p l a y , p. 290, McGraw H ill Book Co. 16. Ins t r u m e n t Divisio n, A l l e n B. DuMont Laborat ories, Inc. , Clifton, N. J. -69 17. Tektronix, Inc. , 712 S. hi. H a w thorne Blvd. , P o r t ­ l a n d 14, Oregon. 18. H e w l e t t - P a c k a r d C o . , 395 P a g e Mill R d . , Palo Alt o, Cal. 19. Soller, C a t h o d e Ray Tube D i s p l a y , p. 296, McG-raw Hill Boo k Co. -70- B l b l l o g r aptiy In a d d i t i o n to the r e f e r e n c e s lis ted, f o l l o w i n g have m a t e r i a l p e r t i n e n t to the the study of p u l s e p r o d u c t i o n a n d m e a s u r e m e n t and were u s e d in the w o r k l e a d i n g to the d e v e l o p m e n t of this thesis. Books Chance, Sleotronlc Time M e a s u r e m e n t , M c G r a w Hill Chanc e, W a v e f o r m s , M c G r a w Hill Cruft, E l e c t r o n l c C l r c u l t s and T u b e s , M c G r a w Hill Gardner & B a r n e s , T r a n s i e n t s in L i n e a r By s t e m s , W i l e y B o o k Co. Glascoe, P u l s e G e n e r a t o r s , M c G r a w Hill Goldman, Trans forma tion C a l c u l u s and Elec trlcal T rans lents , W i l e y B o o k Co. Goldman, F r e q ue ncy A n a l y s i s , M o d u l a t i o n and N o l s e , McGraw Hill G r e e n w o o d , ^ l e c t r o n l c I n s t r u m e n t s , M c G r a w Hill McLachlin, Complex Variables and O p e r a t i o n a l C a l c u l u s , Macmillan Montgomery, Technique of M l c r o w a v e M e a s u r e m e n t s , M c G r a w Hill. P u c k l e , Time B a s e s , W i l e y B o o k Co. Bchelkunoff, Boiler, E l e c t r o m a g n e t l c W a v e s , D. V a n N o s t r a n d Co, C a t h o d e R ay lube D i s p l a y s , M c G r a w Hill T o r r e y , C r ystal R e c t l f l e r s , M c G r a w Hill V a l ley, Vacuum Tube A m p l i f i e r s , McGrew Hill Inc. Journals Bell System Technical Journal electronics Magazine Hewlett Packard Bulletins Journal of the F r a n k l i n I n s t i t u t e P r o c e e d i n g s of the I n s t i t u t e (J o u r n a l ) R. C. A. of R a d i o en g i n e e r s Review Wireless engineering -72-