UNIVERSITY OF

CALIFORNIA

TWO-WEEK LOAN COpy

This is a library Circulating Copywhich may be borrowed for two weeks.For a personal retention copy, call

Tech. Info. Dioision, Ext 5545

BERKELEY. CALIFORNIA

f#

c:~ Iii(0

~ ~

~.°i"-

!Xl :E~ N..,

m.1 a: w ..t 8

It!

a:) en ... 'Bi~

a.

I)x

.!' ~~

,~ i0 !E

oS met ]

~

I

.. ,

Lawrence Radiation Laboratory LibraryUniversity of California, Berkeley

I·'l··.~,:

Cover SheetDo not remove

UNIVERSITY OF CALIFORNIARADIATION LABORATORY

INDEX NO.~.~Ctk-,_9..;.....;:;;;;.:S- _This dccwnent contains~pages

DtJ.Pi"J1~~~~~~:n:~D ;~~;;:P~~~;~~fZT~:;ies ~"m~'iittc~,,,,;~:.q1li'il."l'clM!"l. 1:: I ~~ ., ~ " I 1

If.~m!t~~&~~~i~W!~5::~Wi..r.r1E~l\i?!;~i[::~~)1i;i,,:,,,;;''~;;:''i''~'''''."t::~lJS,1.k. ",. . =>"Xo ,~ ""'" \.' ,. "'''''-~'''''';''''i·'''t'rt';'l'1f1~~~!!1~ C! .~ssrnr)monJ (.' ~1(W m

mmm~§y,n"S';~.,~r!~::::::~ ,::A~'::l .. ',< • l ••

~LASSIFXCAT!O~~~~~f)~,;~c¥.(~,~prk~:l< r:' ::::", { ;;~ OF 'I:::: ""\:'pI" 1 d.~ 1!l~~I~,~C,~.;:no~ Cl,')lWMl1'T~E pr:r':'.~;i! i/'AKING TIlE

mY THE. DECl..A"u, . CHANGe

E,ach person who received this document must sign thespace below.

Route to Noted by Date Route to Noted by Date

.. -------1------.....---+---~ .......-,,f.·i-"-'-_.... ,-----1.....- -_-+ _

.-+/--.....,.--..-,-,-+-------!----------.-4--_.-+__......_,_"*_.........~~.......,.......,+_--....,....---_I_---- ........

.--+-------,,-III--.-..--:"'l-.......---,........--1-......~,--....--.+-......---------t--~._-.......j'-..,__.-_l_l_._--........--__j.-."'"p------4-----r--

----ol---~---I-----4l--------..I--------+_------ -----h--_-..f..-_._---I-I-l -_._-+---..--....--+----,-

I-----..--I-._--,--II-~----I-+--------+--......'--...;.-......'-+--_._--

I-----.....,.!-~.._.. _----:---"'- ...,.+I+... -_.-+..-_-~---+ ......, . _

!i

/\,}

,"

.-<'

'\

..

.;., .......

Special Review of DeduMfied Reports

Authorized by USOOE JK Bratton

Uncll:\s~ified TWX P182206Z May 79.~

REPORTPROPERl.·.-•.~-~D-_E-C-LA-S-S-'F-!E~D.] ,.-". ,

," .'v

\~'

"

l"i

/'

..f't,ngOJ:U:,i.!W J;;f(>.tiDn~"l I~I~;b01°f;ltOJ[·Y

l\'K"m.~Hl ;\;'OZ("O~'!I;l fsp('H}l.fiil Ytf~H\pOnS P)'\"ojectAtorn:h'l J;!:ti.eJ:i("g~yt C(.}mfnil1u~i on$. W$.sh:l.,:n/1:to~1

Batt;oU..;iJ Mom,o;'t':'!t:i.l 1:n$t~it;ntG

Bl"1.)oldlf,\,"iT@l':\ Ni:l:bJ,on~tl I,abol~a.tO)fY

Car'btd(~ &; (1!Q.i'bo~1 Ch~m:t~~li!J (lorpa (11.:=25 Aro~)

Gtll"bid;~ <% 0t11'l:lon Chem.io/i!.l$ Cell"!?" (1,=12 Mea)Co ltnnM (3. lhl.i'iro :("f:d:i;;y (Dunning)GOltlJJlhitt Un:1,vEJ:i:':CJl'by (Fe.:tlla.)GG1/ilC,H"i1.1 nJ1"Hft;r:tt> (~mnp1M\Y'

H9:nfo~od. Di,K'dCJ'c8d Opei'8/lblonsIowrt fH;&,to Co:lJ,(;l ~i@K~311l!~:K G01TJ01:,ati,rm.:LOB AlnwofiJVltVHJI.£,!;hU:3(,)'G),;~j Ixwt;:l.tute of' 'fGohnologyMo:mH~\':d;o Gh'ElIn:I.i}al Compau.:Y'9 D~;y"bon

Nat;:! o:n!\l BU:COilU of Btand~d~

Nmrf,l Rtld.:l.a't;ion I.IAbo:cr:J.tOl'yN:E::l?i~

J\ff,';)w Yod.c Dil"(W'!xHl Opel'ationsOnk Hld1"o No.t:1. onal LB.boratoryPatol1'h AO:irJ. .w~) 9 WashJ.l1g;'"Gon');0 ();hn:tCi~ll Ii1fo~''I!l«il/(;1.on D:tvi!31.oX!9 reDOUCLA :M'eldi(~15d n~;)s(:Hlr(lh Laboratory (11a!"r-au)Un:\v0)t,"RJl to;})' of" GaHi'ol~Klia Radtation IJt\bol'atol'yiJniv<}l<."n i t;y of HO($!w sb'))j;·Wes Jee:c!1 1{OS{:ll."V6 Uni"jfE"ll"Sl ty' (:B':ri~dall)

D© 0 ll'vl () i.e~L (3 atJ. on Pr 0 06 dU)f6Oht'llflis"ta:J' Dapartm6J:').t

hlfo x'ma H 011 Divh ion,lh1di. fit; ion IJa1.JOt:'trt;ol';jl

\

) lh:ll'\i'o:,'t:) i '1~Y' G!i~liro:i&""itl~~

Bt~2:":ko loy. Ordj"fort.d,8.

1=89

10=111,213=1819,~22

23=26272B29<~ Y~~

3:';;··39404104243,,",tfr5464'7·~(~8

49",,50515253=S'!55",,626364~~78

1980c~84

85=868"88~u9'f

98

98

//

~

"

I)

'L·

"

UGRI..=95

Physics=Instrumentatio:n

ArJS'I'HAGT

HIGH VOLTAGE PULSER FOR 184~, INCH CYCr.m~RON ELECTIUG DEPLEGTOR

Qo Ao Korns." Wo Ro BD.ker~ Ro Fo Edwards D and Go Mo F'arly

Radie:tion Lnboratory~ Depto of PhysicsUnj:vGrsity of Califo:rnia~ Be rk(-) ley. California

April 24~ 1948

'1'his paper describes a high V'oltago pulse I':enerator d~voloped t;o

defbGt the beam of the l'A··inch (~yc:lotron at Berkeley.• Cali.t'orni13. o The £l.]:P1aratus

develops a deflecting: Do'benti!)l of' 200 kilovolts thr9t rises f'rm 10 poroent to

90 peroent of peak vG.ll1o in 001 microsecondn 0 The unit emo10ys t1JV'o "lmU",r 100

kilovolt wa.ter cooled pulse tronsformer-s oonnected SVJllclotrica11y ",boUG rTound to

dik:i0ha.rop'ed throup'h the t'/IO turn ,,"ary windj nl"8 of the pulse tl"allSfOT'mocs by

trio:,"~el'ing a batt:)l"Y of 16 paralleled Kuthe 5022 hydroWen thyratronso

"Output; vol tal!es are developed across the 17 turn secondary winchng 01'

the pulse transformer 0 'rho transformer is mounted in an oil filled luc:'d;c, «;>,')0

'chat provides both insulb.tion and compact desil'11o

Contrac;t No 0 W=74,05~Elnt:=48

UCRL=95

Hil?~h Voltage PulseI' £'01" 184=Inch Oyclotron Electric De:f'lectol"

Radiation Laboratoryp Depto of PhysicsUniversity of Californiap BerkaleY9 California

General 0

As is well kno'Wn 9 oha~'ged partioles can be made to rotate in synchronism

wi th a radio freQuenoy field that satiafi.es the cyclotron rosonance conditions 01

In the frequency modulated cyclotron low energy particles are oaught i.nto synohronous

orbits in the center of the oyclotron and caused to gain energy as the frequenoy

i.s lowel'ed at a moderate rate20 In the 184=inch cyclotron9 the starting frequency

for deuterons is approximate l;V 12.5 Me and the frequency corresponding to 200 j:,Tev

is approximately 908 Me 0 Thus each downward frequency swing will cause a l::l\llr'St

of' ions to spiral out to a. large radius 0 In the case of the 184-inch cyclotron~

this radius is 81 3/4 inch?s as can be seen in Figo 10

Deflection or remova.l of the ions from tho i1' circular arMts at "t;he rad.ius

and froquency corresponding to a m.axtnmrl'l. energy of 200 Mev is aoconrolished with

a_ rmlsed electrostati.c u.nit in conjunotion. wi th a. magnetic chnnnel Olf OOtlfoughW30

Haferrinp~ again 'co li'ito 1 9 the high voltage pulse on the eleotrostatic deflector

applies e. radial field that; deflects any ions between the deflector bars t;()1jrards

the center of the oyclotron. shifting thoili" center of rotation enour.r,h to allow them

to pass throug~h the lowered m.ar.r,netic field :i.n the channel or iIltroughOO provided

1Theory of the Synchro-Cyclotrono Do Bohm and Fo 1.'0 FoldyoPhyo Ravo .!~.D 649 (1947)

2Ibido

3Electr-orae.gnetic DeflectOr fo:t' the B~@.!il 01' ~Ch8 184-,inch Gyclotl\"ono Po\wllv Henr-ichv, Kerns~Se'WellD and 'fho:r-ntono BP=138 <Iro b$ pllJ.blished) 0

,.

by the eleotromagnetic deflector9 and consaquentlys to assume a large enotigh ra.dius

of GUl"vature to pa.ss out of the cy'clotron magne'bic fieldo Iri this manner a. significant

portion of the circulating beam of deuterons can be broUght out of. the cyclotron

for studyo

.!:~ll~a_,Timi~~ape~~ld.rv~itu~_

Arl ideal deflector pulse would be timed With respect to two conditionso

(1) The instanta.neous frequency (a.pproxima.tely 10 Me) corresponding

to the radial position of the ions at maximum enargyo

(2) The pha.s:3 of the dee ri'vo);l;a.ge or angula.r position of -I.;he group

of ions with respect to th~ deflector barso

S:1.llC6 the increase in re.diuB of the burst of ions per rf cy'cle J.s apl?r'oxi=

mately 001 :i.nches W'l.d the deflector ba.rs are sps.ced one inch a.part p the pulse must

occur wi thin :t 5 rf. cyoles of the exa.ct frequency on the descending he.lf of the f'm

cyole when the ions are at the mean radius of the detlector barso Since tho f~equ6ncy

modulation rate is only 100 cycles per second,\) 'chis oondition is rather' easily satis=

The second condition roquires the pulse to start at aMme fixed by the

phase of the 10 MC rf dee YO ltage 0 It i3 not n0cess aX'y to fuIf'i 11 this oondi tion

excopt as a refinement since the prob~bility of. a steop voltar6 rise ooeuring when

~a ions are not between 'bhe defleotor bars is. I'slati.vely great beoause of the g~o=

me't;r'iClJ.l proJ;:r;r'ci ons of the defleetor and. of the particle or-bi to

From the fore going it is apparent that the pulse rise time from 10 percent

\

. to 90 percent of full 'Voltage must be of the order of Dol ~J,so The pulse decay time

is to be short compared wi th the f'm. rate of 100 cycles par sElcondo

The electric field between the d6tl~ctor ba~s necessary to shift th~ p~otic10

o~:m'ter of' l~ote.tion enough to allow it to pass through 'che magnetic channel is @.bout

75D~OO volts per centimeter or approximately 200s000 volts for deflector barB spaced

()

,.

'\T(fl J?olf~L~tt:v OJ:l 1;11e J)atle<.rbor B~lr~o::..,_ -c, .~._,_ .• <>'-.~" ''C'-- ,0.: _<- """,,·,,~..-~o- .. ~.:..o,-=""" .or-'""'.~ c....•:.>"'n.~·.".".".....,"",....",·_~.· .. <., """_~_';"~"" '':~r_'"='''''

,(lOXJBi,fj'i;;fj o:{' x'our bars o t'lYIlJ ot"ruil'l,Gh f'oiler! 'Gh(~ posi t1"T6 el(~(~'tr.od8 und. 'hi/W'O 'Lh{,~!,:n(o!Igll'M.:~@o

(;1{Jnell':+tl D;;:I~H~r.i:p'I;;'.o:n ol~ the X'ulsc,n;" 0=:"""~O"_·"'~'''C>''~=·'''''·._'C'~'';:-'''L-",_~""a~~=·,oC'i';'"",n;>;....=~.n>......~_~<l.".:.1.Q. ..,.,..,-O,·~=...-""-'~~,·C>···~r.~£"~

Af'tele' saver~1 mon'~hs of' plfoli.minury oxperimoll'l:ati,0).'!49 majOle' ItJ:r:k"{"l:!:"t was!

t. )I-', "----,'

"

Thus 9 it is desirahle that the defleotor capnci ty and the transi'ormer leakago reaotance

totb0 seoonda:ry~ whon tho oomponents ar~3 assembled9 and not; the transformer al<mso

usedg itt this ease tr10 [l:Itl1<35022 hydrogen thyratr'ons 0 Whfm \/J"Or'king into a. load whom:!

actance ~\nd tho load capncity) 0

secondf),.ry~ much less thfl.n 7 1Jho

o

i two p rirnar;y turns C·O.<1 1.'7 iCiO()Ol,dary 'I,urns • With the resultint~;":;'[':h yol:\:;9.i'·o ~'JCJr turn/J

(approxima.t:ely 6 lei 1,ovolti;i) fla'~ rIbbon t;ypo conductors and iWJlJ.la:1:;ion a:Ci:) found.

·,

, ,, )

UCRL=95

It will be noted from Figo 3 that both the seoondar,y oonduotor and the ~

polystyrene insula.tion are tapered at the high voltage end of the windingo The·,.

,,,-~,./

ribbon conductor is tapered to provide a longer insulation path from the high

voltage to thC!l low voltage end of the winding while the.polystyrene insulation i~

tapered to counteract a tendency of the coil to break down from the high voltage

~nd ot 'the 1i/'ind1n~ to the grounded core e.crossthe boundary between the polystyrene

and the in8ulating oil 0

Two 0011$ are connected in parallel in eaoh trans£or.me:r8 thus making

the total transformer leakage inductance one half that of a single coilo in this

c ..se4 tJh reterlitd to the secondary.. Beca1ase of the large capacity of the defleotor

load~ the capacity of the transformer win4ing itself is relatively unimportanto

P:ri.~y connections to the coils are made through a rectangular coax type

of lead th.at toms an effective 2 ohm transmission line@ providing an approximate

impedance I,\atch into the transformer coils and an effeotive means of trWl&f'~ring

enotgy from the driving circuito This primary lead str1uotu:re Q'lW be obselr'ved at

the low voltage end of the transformer shown in Figo 40 The outer hollow reotangular

part or the struoture forms ()neprimary transformer 1~H'ld9 While a. broad strap9

insulated troll!. the outer lead by polystyrene sheet and mounted on an in'mlating

lucite bushing,p acts as the other primary leado The transformer outputpolax>1ty

is revers.d by ch~ging conneotions to these primary leads o

To mlnmiZe eddy CUrrElnt losses end realize the best available p~lrnlea""

bility for 001 ~s pulses!) Westinghouse 2 mil oriented hypersil type C corea are

usedo Better perJleability would be expected from thin permalloy tape corolS but

at the time of construotion suoh cortHI weN unobte.inQbl~fl The CO:rt'1 ©rO$B=$6rtrfwional

area is made as small as possible,!) in this case 1 1/2 by 1 1/2 inohes v to ~t'Jp the

a:".rage turn length at a llinimumw sinoe the larger the 0011 dhwfiiter,!) thf9 glf'eatelf. I

the leakage inductmee will be.. Interbminar insulation provided on theiIJ<.'> oorea

is adequate for ordinary pulse tran~former8 operating at from 100 to 200 volt$p@~

UCRL=95

=9=

tur'no Howevers in this transformer the voltage per turn is 6000 volts 0 LOS$6lS- - . ~

,I due to intarlaminar eddy ourrants are an appreci able part of the totd li"HHle~

where the interlaminar insulation is inadequateo These losses are reduced by a

factor of two by dividing the core into two 3/4 by 1 1/2 inch saVtiOntsD thus ,making

the vol taga around each of the two segments one half of the total vol tags per turno

Even with this reducod vol tap:e arcing between laminations occurs,!) result:l.ng in

breakdovm of tb.e oil and formation of a fine black carbon powder ovel' 11 'p6:!;"iod

of timao

Approximately ninety perG6'nt of the total power input to the sY'stem lis

eventually dissipated: in the transformer oores as heat. At rated operatJ.ng levels

this loss is approximately 500 watts total or 250 watts in each tra.nsf'orm8ro

Thecoref3 q\dckly reach a temperature of 200 to 300 degrees Centigrade 1:~nd€3:r. these

oonditions D tmd '\vould CfUSe rapid deterioration of the polystyrene insulat:tonin

the coils unless preventative measures are takeno These mea~:nres take t;h*} form

of' thin copper water oooled jackets which surround the portion of. the cora 0.1 wh,jJ~h

the coils resto Precautions are taken so that the jackets 'will not fOY,"Si. ;1:1, /~ho:rted.

turn, and ,:\re adequately i.nsulated from the cores to prevent ShOl"t'ing: o:f' t(H'J (H.):"':"

laminnti.ons 0 If! nO.di tion to the jackets,!)· a la.rge water cooled }food bmou:o.t0d

above i',l.1e (}oreS a f'jirminp.: a cooling: system so at'feotive thnt it op.'3rat:'i.r,.g l~"vel~

the case is barely "'farm to the toucho

By makin,r: the enti:r.6 case of luoi te,!) one end of the case i'bssH fo:rm.~

the hirh' voltage hushinr: and the 'othe:r.6nd is used for primar:l Mnnf,iot,;Lons anr}viatel'

lee.ds 0 The case and the pri.mary lead bushing is vaouum fi lledwith on that hat:';

bean ·de=aerated hy npraying it into an evacuated cheuu'l.:lerD thereby eHm.iKuating a:nji'

',.,

\

)al.r bubbles in tJ:Y; tr;o.u;;fo:rmor Windings that would pormit co:r.ona to exist dmring

fast pulses 0 A transformer thus prepared survived a 300 kJ.lovoH:, output testa

a safety .factor of three 0

UCRL=95

'.

In tHs cireui t!J the switeh must pass a peak nurrent of 5000 amperes per

transformer at a voltags of· at least 11$000 vo'lts with a. jitter tiHle of' less than

0 0 01 ~s in case azimuth timing of thepuls~ is desiredo The hydrogen th~atron was

selected as the only tube cormne rci ally available that would meet theB~ specif':\'0iJ.tioXlS 0

One Kuthe 5C22~~ydrogenthyratronwill switch 5000 amperes for 001 ~s but in doing

so the peak current rating of the tube is exceeded by almost twenty times v resulting

in hydrogen cleanup and short tube life 0 By increasj.ng 'bhe number of tube,:l to 8

in parallel on each transformer or 16 in all and introducing a v~ry small. inductanc@

in each plate lead to make the tubes 'share the 10ad!J it is possible to sv,ri't;ch the

entire 10,9000 ampereso To further insure that the tubes will share the 10ad.9 t,hey

are individually m~\tched and each two tubes discharge one oapaci to.r 9 as shown in

Figo 2 0 'l'hase tubes cannot be operated at plate voltages above 119000 '\Tol''t1S c even

though rated at 16 9 000 volts» because,. in operation, the plate voltage :reYEir'sad in

003 118 causing arcing between plate and grid and tube failure at platEl \1011:.118;01$ i,;).bovl;

119000 volts 0 Kuthe Laboratories havl!l developed hydrogen thyre:tronswJ:l;;h 01. much higher

awi tehing capacityo Their H=6000 type will replace eight 5022 a3 in thiB o:u("cmH butD

unfortunat~ny~ the tube is subject- to the same plate to grid arcing ,troub18s8.S '\:;)16l'

5C22 as and the same plate voltage limitation existso

Special firing; circuits have been developed to fire the 16 r,a:f?:il101ed 5G22. ."..

hydrogenthyr!9-tro'ns With~jitte~ time of less than OeOl ~so Ref'erring to Flgo 2 9

it will be seen that each tube is connected to t he trigger source through ~U1 artifioial

transmission line 0 These lines provide a 1000 volt trigger of' 20 ohms imp(~dJ;mCei

" for approximately 0020 microseconds 0 Applying; this trigp:er to 'che Vrids results

Vi in positive ionization of all the tubos in 001 ;it 0 001 p,so Thus frxr:o no dLC'J:':'LCll}:t;y

has been experienced in firing groups of up to 16 tubes simultaneouslY9 }Jl'ov5-ded

that a low impedance trigger souroe is a:vailabla to oharge all th@ tran~=

" ,

.~ ,

UCRL=95

=11=

mission lines in parallelo

A satisfactoJ:'Y pulse storag:o ce.pacitor must meet severa.l requirements 0

Firsts 'the energy storage must be large enough to char/l:e thf> deflector ca'iEl,city

and provide for other losses in the system, in this case~ 0005 Jl.f per transfo1"J1l:}I'

or 0 0 1 ~ in a110 Seoond, the voltage and peak current ratilJ-L;S ~ust be a.t least

11,000 volts and 5000 amperes per transformor rpsn€lct:l:velyo 'rhJ.rtl 9 the {}apacitor

must withstand a complete reversal of char[':e in 0 0 3 Us 100 til,les a s800nd wi.thout

failureo Fourth, the internal inductance ~hoUld be a. small part 0f the total

allov/able leaka.p;e inducte.nM referred to the tre.nsformor primary '( tota.l allowableI

inductance referred to nrima.ryis 7~(l7/2)2 or a.pproxima.tely OoltJ.h)o

All typos of capacitors obte.:i.nable ,wer-e tented,. none \,rovin'~ ;>:3.tis.faotory

in all reS1l6ct.s 0 The lowest .failllre rate was obtainod with f. Oo()0 \1,[", 16 3 000 vot!:

in l;aralle1~ or f'Gur ner trans.rorme:r~ They are completely irnrnel":;;ed in oooling, -,

water to remove the heat venerated in the dieleotric withth~ ra.pid c,harp.:1':l lr€lyersal:so

The failure rate on the~_~'3 units operated a.t 11 9 000 volts nla1.;€J 1Tol'tD.f"6 has been

one for every 10 to 20 hours of service 0 Another lIDdesirahle feat'J:t-e is th6 high

-internal induct,enee of the capacitor g in thi? oas6, a.pproAilll~telJ ~:\Q13 ~tho ThE!

effect of this inductanoe on the oireui t performance is discussed lo.tsl'o General

Eleotric is developin~ a. snecia.l capneito:r6 to remove 'cher-;El two undesirable condi=

tions"

,.;J~

'" comJ.aJl'!i.~I'-J1n p

-~'.."">#'","",,,,-----

5' ,General Electrio Oil capacitor, Cato 26F380~ 0 0 03 ~9 16 9 000 yoltso

6 "General Electric has installed water cooling in So capMito:r ( 0 05 ~9 1.6,9000 volts~

Cato' 19F104l that solves tho hoat removal problemo Th6;)r a.re now vlOrkinr': on tho ')roblem cif re ducin~ the inductant;<i be low 001,3 tJ.h 0

UCRL=95

'.=12=

was satisfaotory.elec~ical1y and needed no cooling~ but was evant~ally replaced

,f because of the bulkiness of the 6000 feet of RG 8 U cableo

The pulse discharge capacitors are char~ad with a DC resonance charging

system using four 304TL tUbes connected as diodes o An interesting variation from

the usual DC resonance charging system is the fact that a plato voltage of IlvOOO

volts on the thyratrons can be maintained with a power supply voltage of 2 9 750

volts$ a step up ratio of four to one as compared wi th a maximum of 'two to one

in the usu.al case 0 The reason fOlO the higher step up ratio is found to be in the

vol tago reversal of. the pulse discharge capacitors each time the system is 'pulsed"

An analysis 'of the system reveals that the step up ratio is dependent upon losse,s

in the entire s~rstem~ and exporiments have shown this to be true 0 step up ~'at;ios

as high as ten to one have been observedo

A photograoh of the entire system is shown in Figo 50 All t,},6 CO:;l'pc:n.cm't,s

are assembled as compactly as possible because of the impo:rtano61 Of' h'd:;,pit:g 'che

primary,oircuit inductance at a minimUIno The completed e:ssembly is loc/;:,'t;'):id. N:U;h

respeot to the leakage field of the cyclotron so that thethyratrons are in a mini=

mum field region and the transformer oor6S 9 while in a stronger fit.~ld l;~egionD moe

oriented so as to be insensitive to the stray fie1do The assembly is enclosed

in a copper lined housing to prevent broadcasting of the high vol ttlg'O pul.ses in:lto

sensitive countin~ equipment nearbyo

Performance D~tao--r(...

r

since the main object was to oomplete f. working'um. t in I&. rtlJ.nimUIn of timE:, 0 Some

measurements were made 9 hOW6lVEH"v the most use;f'ulof' which was the deroomination

of the output voltage with respect to tim.eo This curve i8 plotted in Figo €L

.''f

,', .. UCR!F95

-13~

It will be observ~d that the effact~~,:rise tilne is about O.. i5 tJ.8» 50 percent more'

" then that desiredo The major part ot the iri'e:tease is felt to be due to the i.p.ternal

inductance altha capacitors usedo Sat1.sfactotYpertoraance h" re.ulted. even'

though t~e rise t11ne18 appreoiably longer than that desired ..' , In general it 18.

felt that increasing the peak voltage ~ompens..tes fot the longer rise till.' of the

pulse ..

Oonclusio'no '

The equipment, while meeting the present' needs on the cyclotron!> 'is not

, the best obtainable", and a superior system. is being developed to r'eplace 'ito .' It is

hoped that the new system will abo make detlecti-on of the proton beam posdblewhsJ:1 "

~, '

, "

....." ,

, .BW!5=20.i48 ' ,'" ,',Inf'onna.tioli DM.1oft

.~ )

.. ~

~-

I

--..--~ '\

.~

200' GROUND

A A

~o

I. II

I/

//

//

//

J III LI I }J JJf

I r::t=l--l+1-- ~ i­-~ L.+f __t.;J I I

I 3" -.!- 200 .~~ 1"P~

ELECTROSTATIC t L=.t-t=t--!~DEFLECTOR . J III I . 200

I r ~J.1f

311

77"4

.."

.~

A

A

------

+

120 0

HIGH VOLTAGE

PULSE

GENERATOR./

III81­

4

\\

\" ORBIT OF~~DEFLECTED ION

.............

...............

--..........--- - - -----

..MAGNETIC

DEFLECTOR

EDGEOF DEE

:t • 'L \.--

r~~I . •

.~

UCRL-95

Figo 1

Cross Seotion of Cyolotron Showing Defleotor Components and Path of a Defleoted Ion

D'Ch.-95

HIGHVOLTAGE

PULSE

TRANSFORMERS

~--+

IiIL

IIIIIL

PULSE STORAGECAPACITORS

TRIGGER.---t~PU LSE

LINE

75C22 1S

75C22 1S

. 304 TL. .CHARGING

. DIODES

TRIGGER IN __-'.J

-+H. V.

POWERSUPPLY

Fig .. 2

Fast Pulse Generator Schematic

UCRL-95 POLYSTYRENE

..~'-<.. ~

~'LEAD

3 11

.,

. . ... ((I

~' :;;..=7 ,r,-r........:tfr.i:..{ i' I I I I

, I I ' (If I II ( ~I ~ I I W I I I I I,// r~ /11}lyl{)I't t1/~ I I ! I I I,., II IIAIV I'IIA/rl l l I I I'

IV/ / / / A tJ ~.u 1[/1 ;II r4I y;Ill' f. I J,..~~tj .¥f¥if.{,YJJJJ.J.l-

~

Cu LEAD-

PRIMARY WINDING:2 TURNS 0.005 Cu FOIL2" WIDE

INSULATION: 3 LAYERS0.005 II . POLYSTYREN E RI eeON

17 TURNS 0.001 AI FOILTAPERED FROM 1/2" - 2 II WIDE

SECONDARY WINDING:

J-

IilL2:' 2" _I

Fig .. 3

Coil De-ve-loped for 0.1 tl.S 100 IV Pulse Transformer

ITZZL1 cu

ill] AI

o POLYSTYRENE

.,