Electronics Meeting - Brookhaven National...

transcript

GEM TN-92-239

Electronics Meeting - Brookhaven National Laboratory

Session Chairs: M. Shaevitz, D. Marlow

November 5, 1992

Abstract:

Agenda and presentations of the GEM Electronics Subgroup Meeting held at Brookhaven National Laboratory on November 5, 1992.

J:: .l..11:::J>..4'..'••• .. 4<;;.L•J .... ='_ ... __ - - - -· Collaboration Meeting at-BNL

Thursday, November 5, 1992

S:30 Introductory Remarks

antral Tracker

Speaker/ Discussion Leader

Harlow/Shaevitz

8:35 Si Vtx Electronic• 8:45 IPC Electronics: Status and Front-End Simulation Study

Mills* Muaaer O'Connor 9:15 IPC Preamp Statue

alorimeter Readout I

9:30 Update on Dynamic Ranqe and Zero•auppr•••ion ltudJ.•• 9:40 LKr Preamplifier/Cable Driver ltatua 9:55 LKr/LAr Timing 1'asult• 0:10 FEP-baaed Level 1 Tri99er Statue

(Coffee Break)

roint Session with Muon Oroup (propoaed)

.o: 30-12: 00 Comment• on Muon Triggering Electronics Coate Diacuaaion

L2:00 Lunch

;alorimeter Readout 11

1:00 Status of ADC Readout Work at Nevia 1:45 Cal FPGA Status

Readout Requireaente for KCAL

~on Electronics (Electronics Group only)

2:00 Muon Front-End work at BNL 2:15 CSC Readout Requirements ' Option•

System Integration

2:30 Early results on transformers in magnetic field• 2:45 Groundinq and Shielding in GBH

3:00 Virtual Level 2 Triqqer/DAQ Scheme 3:30 Diacuaaion

Plan for TDR

3:45 Work Aaaiqnmenta ' Schedule for Drafts

Plan for Coat ' Schedule Work (Optional for some)

4:15 Overall Plan for GEM 4:30 Electronics Coat Estimating Plan ' Work Assignments

rrabya IWacia* Takai Croaatto

Shaevitz tbd

Sippach Gara tbd

O'Connor Wixted

. Freeman Lau*

Dorenboach

Marlow/Shaevitz

Fischer et al. Marlow · · ·

5:00 Adjourn (the meeting may run somewhat later than thia, but it will ... _ • .:_.:""-'·• ... ~ ... 4.h h .. , 41\•nn PM\.

Presentation by:

S. Hahn

' ·S:-. ln.A~ ~CC,...w~~ ,_,.(~'' "'; ,. 1 ' .

MCNI ·· ·, '. · ··

' ... ~. UTM(,, Ho,.rl$: tiw...0.· ~- ~J;.tt-l ·. , ' · Are.'itee"CAre. 11e1111 1.Ule.(I :·

. . .SwbC.:r°"4,.ts '-'1' sfm"lo~' .

LAAll-/:JllLA/ LU.';'l,·c. CRAl>I( :.+tt,'ttl ~> -~on . ·; ~

· '~i f"'"':r- · J'r.e.se,.ttAtl•" .t, Yell.ws-1."t. .. ;

.,'· ; Pos+.t s.essro-. .:t r••E -IJ~S ;,s ,.T ~7 C.•nt,.J- to 6e s1"w4/-t;le~•1-.~l I z e,,,, vs. 1 r e...

. '<¥.:: -.

. '' .

. ; ' t! •r&t'

Lee.a. n.~ of ~nel~

/V4'l1e.

c..,,o.,;t.~~

htcttAt lfr;p rY#is1'.11~ (-tl'Oltf$h1i£St.l'l -t,•>te, .e.ffttJ)

~«.·C•~r~ Reo.L..4.tt:c.. ~ e ~fe,,./Jl

77,,.~- IAJA/ I< .a:•

r ,uttAJo &....-cro .ss;''!I elc~J,,, "1

••

-~~· ~-i:· T. ·.;.;..-~ ;_· ·.;....· ,.;;;.. .. · ..;;i..~~~::;:~-· v(}..r iid c. r· .. ·.,.·. , •••

"'.c·'' ~--t.(·\••• e • f • ..... , .·':'· -··!· .. ·:{. . , ~-~)'·-· ... ,· ... O.'-fl,,.. l. •. ,,,,. ..

~- T"'""'"" I I I

,.,,, .a76/

,r .....

" Custom Bipolar Integrated Circuit Designs for the Front-End Electronics of the Silicon

Tracker Subsystem for SSC-GEM Detector

Sanakoo F. Hahl} and Mauroon M, Cafferty -I Loa Alamo1 National Laboratory, Loa Al1mo1, NM 87545

Abstract

Bxlnmtly hllh rlllilatlon IN!OklrDllllll (up to I ~ aver a 10 yNr llltllmt). a larp .number of ob111nt11 (3,2 ld), hl&h lpMd. low nola rtqul1'1"*111 (1;5o •INtrtRI l\?yfl), AlllllOW poww co-p111n (1 mW/obanMI), pr111nt a Alrm!l!IW• chllll8" fir ~ lltolrlnill of 1119 1!1141n ~ .-.,111111 Air lilt llC.cmM (Olmma, !ll~n, · Muon) ~. llmulatlllR f'9111111 fmll tbl \llpolflr ""~ fir 811 lllllllq l• .. p&ltll ol~l 11 Pfllllll ~ .. lllJUl!l A'llm 4U!~ll llllp _ 4'tloien ani prueatld Ullnl A'PAT 1!1111 aimta llllllllllVr mo4ela.

I. INTRODUQTION

The decillon Wlll . nwt. •rller that a \llpelar to leehnolol)' wu 10 111 1iHd tor Iha prOl!lllllll ol tht ..... 1lpal1 llvm Ille ll"ltrlp dtteeton to 11!ff1 1119 nolll, 1J1-.c1 and poMr requlr1mtm1, Tb• m1ln lu~n of ll!t •"'1•r .IQ 11 10 llllllit 1111 llllllllllll\IOUI lop9 cMolalon of •lthtr 111! or nohit per miry b9am-oroulllf (with a perlocl of Iii ns) tor_,,. ~hannel bued on thl ~1- of obarp colleelecl llvm ·the strip dotKton, piat ID Iha ll)W CJC:CNP&ni:y of 1111 lhan 1 %, the dead-lime of the clroull operation betwoon minll oan be 11

10 the 90mpar1tor. Without the ratorallon, the drift of the D~ lfVtl WDllld pl'lllRI a NrlDlll oontrol pniblem lo the lirGUI\ 8f11J1\IDll (Plpl'I 2). . Thi IMMllplrlllllr clNlgn 11 RrtltJl!lferward wlt!i a illir,NAllll•palr lllpvt 111p followed by GUmnt•mlrror ampllft~ ftlr 1111mn1 euqnn (Ptaure l).

~lltOM lllMIHO

. .• 1, ';

Q~ Ill

lltll Ill lltll UK IK IOO

· much 11 i'O RI. Tho dulgn featuru and the ml\lor simulation results aro deacrlbed. ,

Pia. l Prlllllnp .t: ln~nnedlate-lllgo amplifier

. II. DBSIONFBATURBS I

The preamp U101 tho familiar cslCOdo conft111rallon tith an Input transistor 1eometry of three llmu Iha area of the minimum .,Ometry traneWiin tq rodl!G!I the not .. ffclm. the b8lo spread rul1ianc;t. 1'11!1 ~ rulstor Ind ~t9r 1et the 1haplftl time of 30 RI. The t'Mdblck rulllOI' I• made 11 larp 11 poulble for minimum noleo conlrlbuUon while nmahila1'lmlll •llOlllh to maintain ,_nably oonstant bias conclltlo111, which IN Hlllltive to transistor pin ~. ovtr the liletl11111 of tbl 1)'1tem. ,

1'111 lntetlllldla~ !llllPllllfr 11 a minimum ~mpoillnl 111119 which 11 very Nlllltlve to the bias l4V9I 11 the base of Q22. The rulllDn, JU and R6 11'1 acUUllld ft!r maximum ' poAlbla pin without 11111hln1 Q22 lnlD 1111n1lon. This maxlmlzu Iha 1lpal to nolte ratio from lhl1 lllA (Flpro 1),

The 1haperlbulllne rutoror u,. has a ciOMd 10\)P pi11 of about eJ1h1, lor the 1l1nal, with a 1111111 111'in1 of llboul 40 mV for a ~ 11tctron Input char... The blloline rcstoralloo cl~lt cs""' IOD19 nol11 dl1radatlon but has a crucial role of 1lablllzln1 Ibo DC blu lmil of the Input node

FRQM BIASING NElWORK

R21 2.2K

IAIELINE ~l!ITORER

Pl1. 2 ShapoMmplUlor

... ..

.. ...

••

......

... .. °' • ••

ft•• = .. ..

,.,.IT ITAlll

ftll ...

._.... '' I,

... ... ...

' ' ' JIJ. 3 lehlma1l11 for Nllll~Om ll!)'Olll lllilizlna lltl AT.,. At.Al JO &llf ~

Ill. Sn.rut.AnONRBsULTI

A. AT&TTileArt'Q)'ALAJJO

Simul1tlo111 were ,,...rormtd uatn1 model• ft'om !ht AT.tT CBIC·V PfOll'I', A dtllp lrlJICI on the ALA! JO Ult array modtlt will. bf lllbmlta..i IO ATAT for 11mloQUllDJ11 fabricalion, and Ibo chip wl~I be teated wbon avallablo. The schemalio and lhe plnoub. aro thown In Plauro 3. Tho Ille array 11 for 1mall-tcalo 1pplloaU0111 and conlal111 minimum area geometry 1ra~lston of 32,5 x 47 .5 11m which aro IOo large for tho final design. The final design will be based on smsllor geometry lranslslon.

Slmulallon results ualng Mlcr0Slm'1 PSPICB were obtained for charge lnpula from . 6000 ol~ron1 to 48000 oleclro111. The c:urrenl 10urce 1lmul11lng Ibo waveform out of a 300 11m lhick PIN diode (lJ 111hown in Plauro 4.

1.a,. .--.... , .... ...--.-...---...... - ...... ___ _

1.0t;I --- __ ,____,

I ICICUln -.- ---+-·--- --....... . .U.::.--. -- ~ ·-

Simulation1 were run for llrip lengths ot 6 cm, I 2 cm, and 18 cm with nominal lranlislor current gaiDI for beginning of

lift ' 1114 lew ~ plfti '9r ..... Ind ., lit. Mf11111114M, 1'ht llrlP dltN!or. WU~ llllllf lumped ~ willl Q!ffl ll!'IPI and tiv.o ampllft~ In: pirall,I u lhown Ill Plpre ' 1a1 10 ~ ~illina elftlotl bolween • .i1....ft. !lblnnelJ, ' ' ' ·, . ...,,..,,...... ·''

_..__NMATIOM9 1111111 mW1n10 ... 1011i>litcaWAL1NT QllJCUlf l'Oft ~I OM 11C1D4 ~---llC'lNllllCll

Piauro' I . ' , ' ..

•111111lallon ruul~ between U11n1AT&T111bclrcult modol1 for tho l'OllJton and oapacllon and u1tn1 Ideal nalltOn and oapaoiton won vlrtllally ldonUoal, but tho not• perlbnnanc:e w-- !, aliout I~ wont for -~ IUbalrqul\ Ulil&On and oapatjlOn· Ti1' 11mul1tlo!1 11- for . !'1rfl! lllllplillen 11Ulltln1 all '""°'roult lllOdol~ were l!IUCh lonpr lhan P~lllll, ~. Ideal. n1l11on aiid capaolton Mn lllld In all mbleqll!lnt re111lb. . . , . ~cal llanll• at tho Input ot tho r:ompuator to~ It c:m

llrip length, ... , end lqJocdon are lhown In Plsure 6 while the comparator oulpllb are 1~ In Pll!lf'O 7. Note !hit tho time-walk i1 lllCUllred at the filling edge of the comparalor output waveform to take advanlago of the plCUdo zerocroning lime pick-off elfecl.

... ·-·-· ---~-··--· ... _ .... __ ,._, . ...,-~·'1-i··-·-· ....... ~ .. -····--·- .... ~ .. , .............. .

Plprt 6

, Flgure7

· ~ 1lmulatlon l'Cllllli are 1Ummed up In Ta!>le l for · beglnnin• of lifetime •DC! ~11d of lifetime. Small llignal amplitude, ti'™! walk, and noll!l r,111111 •"!! 1hown for 6 Cm, 12 cm, •ncl 18 cm detector ~trip l•nJlhl. ~ amplltud' ~ -•ured for tho 60!>0 tloctron i11pu1. Tim, wal~ wu meuured for 6000 lo . 48000 electron inp111t .. with ~ comperalor lhrtlbol4 !mil llCI to 60% of ~ •mall 11Pal ...i.., for each ~r llrip l.uth. The 1111111 wa1 calculated at tho lnJllll to tho comparator. · ·

. .. B. Harris UHF

Slmulallo111 ware alto performed U1in1 mod•ll for Harrl1 UHP •mall pomatry ll'lllll1tor1 for J 2 om 1114 II om dotoctor 1trip lengtlla.

The nollf for the l2 cm fllr-end wa1 2''40 ellqlrona ~ 2210 eltctrou lor Ille near-end. TIM 1lma walk - 13.7 RI

· for Che tllr,.cnd and 10.2 na for Ille Dllr-cnd with tho threshold level equal to .5000 clectro111 (no 6000 electron • triqcr).

The noise for lhe 18 cm far-end wu 3550 electrons and 2710 ol~on• for the near-end. The time walk Wal 9.1 hi

for the fir.end (no 6000 or 12000 electron trlger) and 9.0 01

(no 6000 electron trigger) with the threshold level equal to 62'0 eleqron•.

Table I

IV. CONCLUSION .. A preliminary design for the front-end electronics for the ·

Silicon Tracker IUb.Y,.em for the SSC-GBM (Gamins, Blec;tron, Muon) Det19!or bu been 1lmulited and will be IUbmltted to ATAT for 19ml-cu1tom filbrlcatlon.

Tha reaulll 1how eiccollent beainnlng-of-life perfi>rmance and ac:eoptable end~ I~ performance (10 year lifetime, 5 MRad total dual) fbr required time walk (<16 na), with the given power (I mW) for a 12 cl!I dotector ltrlp length, but clearly d•gradtd performance at 18 cm lengtlui .

V. REf'BRBNCBS (IJ W.C. lallor, H . .J. Zlock, W.W. Klnnlaon, K. Holzcholter,

"It Modal for tho PerformaDOI 11f SlllOlln MicrOllrlp Dtllolon•, N"ol•• l"nru111•nl1 and Al•tllod1 In Pllyrlcs R•ntwh, A303 (111?1) pp. 285·2117 ·

(2] Iuy Klpnl• "An1la1 Pront•Bnd BlecitroniC. for tho SDC Sllioan Trac:ker", Warbhop on Pront•Bnd,B~niCI for Blllcon Detectors at Jluturt Hlah Lumlno1i1y Colllders Y •llow1toD1 National Partc, Wyomlna l'7 Stptembe; 1992 •

Presentation by:

Jim Musser

. t>-~··all r-o."~"u (j ,..,.,,.,.~ s !ft:~~ () t.a. Re.' .,.t ba ~ow..\~

-1'·:~

~e A~ t>~~ar\. r-e.\): .. ~ tJt\) io/fl.,.e t,, t>"o~ frol>L.fW\ tt.60/ueJ... . · ·

F~t:>c.: j.IA-'looP r .. ~t,,. o~-+-..A..l ~ u. .. ~. c,,.•. LI l ~ ..... , c."' e,,., ~c.r ooe .. olt#fl.

READ - WRITE ADDRESS GENERATOR

- FF register based status array

Indicates write protection of an address for readout

Two status bits read out per event clock

Status bits control skip logic to pass over protected write locations

- Parallel adders

Generate skip addresses

Status bits select proper adder output for next address

Write address latched for external analog memory and internally for reading out next status bits

Read address is delayed by the trigger calculation time and is latched upon receipt of a trigger signal

Read addresses are pushed into FIFOs for eventual readout

- Status of Read - Write generator

Schematic capture completed

Functional unit delay timing completed and passed

Optimized routing in Xilinx IC in progress to meet system level speed

Scheduled for completion by November 92

Aolcl 1,4, o.ncl 7 .

;' ...... 21

Select 1,4, or7

;' "' 7 /

FF register ck ' / 7 / 7 / 7 ........_

/ I / / slice write tlMe

write sto. tus o.ololress FF row/coluMn clecooler

I/ v 16 / coluMn 8/ row

I QI +' d 2-8x8 FF write 16/ BO +' sto.tus o.rro.ys VI I /E s... +'

/ +' /

Bl SJ. 128 VI :J 8/ +' 0 ~c +' VI

I 1oclk

• 4.· •

AO l. '.

S/RS I 0

CLOCKS 0

MC LOCK 0

I WR2 1

I ' I I I 3270.2 3154 3203.8 3253.6 3303.4 3353.2 3403

READOUT CONTROLLER IC

- Three functional blocks

Read - Write address generators

Timing And control

Data readout

- Prototype technology

Xilinx field programmable gate array - Static Ram based structure

Three chip solutlon -Two Xlllnx 3195-3 ICa & One SAAM

Xilinx 3195: 50 - 80 Mhz system level speed

Great~r than 9,000 nominal gates per 3195 IC

- End product technology

LSI Logic LRH20K radiation hardened series gate array

Guaranteed total dose specification 3E6 rad (si)

Silicon - gate 1.0 micron ( 0.7 micron effective) 2 - layer metal HCMOS

Timing & Control/Data Readout

- Timing & Control

Flip - Flop synchronous state machine

Synchronous with the event clock

Generates all 1ub-clock1 and control 1lgnal1

- Data readout

FIFO data buffered from the output of. eight Flash A·D converters

SAAM stored thresholds are magnitude compared with FIFO buffered data for zero suppression

Non-suppressed data are pushed Into a FIFO for data frame packing

Non~auppressed data, read address, bunch crossing ID, and frai:ne sync word are packed into serial data frames

Data transmission is synchronous, frame synched communication over' fiber optic link

- Status of Timing & Control/Data readout

Block diagram completed

Schematic capture In progress

Scheduled for completion by February 93

t\:) fo-6

Aclcl 1,4, o.ncl 7

Select 1,4, or 7

FF register

write sto. tus FF row/colul'ln clecocler

16 J c:olUl'ln B1 row

I ' ..

slle:e

d I I 2-8x8 FF write +'

" sto.tus o.rro.ys .L !.. +'

11201 " 18 ::1 +' d +' .. ---'

bc:lk . 8

4 I push sir oe o.srt pop c:htl

Aclcl L4, o.ncl 7

..---1--1 Select 1,4, or 7

FF register ck

Reo.cl/'Write Aclclress Gen. 11/2/92 Rev.3

7 1 , •push / 7 bit x 6 cleep FIFO

reo.cl status FF row/colurm clecocler

161 COIUl"ln 8/1 row

Li .. +' d I I 2-8x8 FF reo.d +' sto.tus o.rro.ys ..

IOI!. !.. .... .... .ii I 128 •

BU~ d .... .. ---'

FF register

reo.d tlMe slice. sir I o.clclress • :>

subtro.c:t 1

1-+---------.1 reset

push set vs et

7 x 3 deep FIFO

subtro.ct 1

i---------_J set

set esr vset bclk loo.cl c:k ep/s vo.llcl c:onvertl. trigger

shift/loo.cl TiMing 8c Control FADC

usn srd swr w-do.tci zero ID e vo.llOI vo.liOI vo.licl ~~n~

F"AD 1-8

ADC ush

I\:) N

d fpop ~ ---~ 0.

I do.to. VO.lid

rec.cl ti Me slice

sync word

c A>B=push

I bunch

counter /8

pe3 8 / v do.to. OU

dpop 3

8/ 128 x 8 do.to. - s srd

t / SRAM

FIFO w

t 8/ swr e / g

I pe2 •

3 8/ V do.to. In zero VO.lid s

shlft/lond t

Po.r. valid e

I / . to

To F'lber /8

3 serlo.l optic

c:lpop x-l"lltter bunch - s

F"lf"D t p/sclk e

pel ' Do.to. 3 Reo.dout Bunch c:lpop

& --- s 11/2/92 't sync

0. Rev. 1 F'IFD t e

sync vo.llol peO

Will f ••'""''"' ~~

IPC Front End Simulation

• Use GARFIELD to simulate IPC

4 mm chamber thickness 2 mm wire spacing 20 micron wire diameter 50-50 CF 4 C02 2700 V operating voltage

• • .. • • u

E Field vs Position • u u u .. .. .. ..

Drift Time Contours . .

.• ... :.:~ .

•• ·----~---1 . . .

•• .._._.~~.......,..<.-~~

I I C t I I C C i I °C I cl

• .. • .. .. • " • " • u

• • .. ..

c c rt 1 c c r 'cc ct ct c a tr t ti, - . .......

• Drift Velocity vs E-Field based on IU measurements • Diffusion - based on pure C02 , CF 4

Drift Velocity vs E

Diffusion vs E

I ~ f: .. ~ ..

" ... u u w .. .. u -·

ltCtllCllCC : Lea C:4'J SEIP ., IJ't:lll llt"'>

'-'---~~~~~~

1111111111111111 .., 111/1'1 £11/1' •• •Ion: torr].

• Simulate signals in IPC includes random trajectories

diffusion Landau fluctuations

-7.5 .

-12.5 .

- 17.5 -

w............._,_._.._,_.._._._.._.__._.__.__,_W...W'-'-'-''-'-'--'-'-'-·l_._,'-'--'--'---'-'-'-~ 0 50 100 150 200 250 300 350 400

(ft•>

-·..!·· - '· •. -_ .:~ . : ": '.

• Simulate signals in IPC includes random trajectories

diffusion Landau fluctuations

0 50 100 150 200 250 JOO J50 400

_:.:_ ..

IPC Front End Simulation • Apply transfer function from preamp/shaper

Based on measured response of prototype . , amps. Pole Zero has been added to eliminate ion tail.

Amp. Impulse Response. Function

I pC input 3000 r

1' 1000

-500 D 20

~------,.,../ _L

BO 100 120 1~0

amp response

Output pulses 2000

o_J I ' I , I I I , I I I I I I I I • I I I I •.....J...J......::r::=.,,.JJ

0 20 40 60 80 . 100 120 140 160

wire sinnol

0 0 20 40

29 . WIRE SIGNAL

• Choose position on pad at random • Use position to make three pad signals according to

known induced signal distribution. • Sample three pad signals at 60 Mhz. • apply 3 types of 'smearing' to samples

noise - added to sample usina normal distribution time jitter on sampling points shaping time errors ·

IPC Front End Simulation • Electronics Noise

Look at three samples - (3rd, 4th, and 5th )

Approx Sample Loe.

2000 I~ 1600

1200 s .

' ... ,: _ _.:: ·. ,1. . : ·-:~ ·.·. .

a ~"°'L.lr. • ~"° ... 1c •• ,_. t".L~~ .,;4&f\e..\. · al:"'.

i':4 "'..,...,~ '·. s .. ,s, i• 1c_...." •ft.a...cul c.~ .... , .. al:.,\. "'· td )(.

~ .. -a.c·~· S', .. a.s "'"' · s, s .... ~ .. s-1

4th sample x (fit) vs x( actual)

ID 1000000 " ENTRIES 965

O.OOOE+OO 0.000E+OO 0,QOQE+OO O.OOOE+OO 965. :Q,OQ.OE+OO

2 - O.OOOE+OO O.OOOE+OQ ;.p:OQOE+OO .....

~ . ~f~i~·~· ·. ~

....... ).:-·. >- t'·

.~ >- .1: ~ /' >- ·'i···

·~sf· >- It'

,,.f.I. ..•. " ,,.-r~

,:..;~'

~ -i~ f: .. ., .. ,.

>-_;j'f:' .. ..;~'i:·'

·.->-__ .,t/"i)y.·.

-1 - ·.,:ff' ... ,>;

:1! . .J.l.

>- .>~·:-:,_'1·· .. . ... . I ~· ~

.. •.• .. . , -2 ~ ·:;- ·.::--.-

" .. ..

->- . . • • I . . ' I . • • I • • • • I . . . . I . . . •

-2 -1 0 1 2

SXz4 VS XzlN

4th sample x(mean) vs x(actual)

-2 -1 0 1 2

SWXz4 VS XzlN

• resolution vs noise level

I ample 3 Rea olullon

3rd sample

(l.•t.•li) •IN•V

IO •

11: • •

0 1 2 3 "' NolH (mV)

Sample 4 R eaolullon

"'° •

4th sample

= -;-30 • i· Cl g 20 : • " '& 10 • I! -

( le<tS,•S,) I ¥81MV 0 1 2 3 4

Nol•• (mV)

Sample 4 R eaolullon

5th sample

( ··--·~· .,) : , .... ti

Ii •

1: • •

0 1 2 3 4 NolH (mV)

IPC Front End Simulation • Noise vs Resolution

Average 3/4/5

40 6 'Ci)' 30

·.;= c ·.: e 20 0 u U) ·s 10 ~-

Sample Ave. Resolution

• .. ,., -

• •

Q+--~-1-~~+-~-+-~---"

0 1 2 3 4

Noise (mV)

IPC Front End Simulation • Resolution vs Time Jitter

Average 3/4/5 No electronics noise

Pos. R esolutlon vs Time Jltt•

~ 60-g 50 ~ 40-'e 3o--, 20->-

• •

~ 10 • ~ Q-t-~~:f--~-+-:~~-+--;~---1;

0 0.5 1 1.5 2

T lme J ltter (ns)

IPC Front End Simulation • resolution vs shaping time errors

No electronics noise No time jitter

Pos. R esolutlon vs S haplng Error(%)

80 •

• -U) Ln._

• C UV

U) = Q) t3 40·

a:: ·- 20 E ,_ - o---·----~·---------1-:------~:

0 20 40 60

S hoping T lme Error (0/o)

Sample 3/4/5 average

IPC Front End Simulation • TO resolution

TO= (sl + 16*s2+32"'s3)/(sl+s2+s3)

TO res. vs noise

TO res. vs shaping . time error

TO R91olut.IM

I 1.04.,

1.03·• • '

• • • 1.02 . . . '

0 1 2 3

Noise (mV)

TOR esolutlon vs S hoping Error C°fo)

";'4 .s. 3 • ' ., 2 c! 1 • ~o •

0 20 40

S hoping T lme E nor (%)

• TO vs time of first electron at wire

ID 1000000 ~

ENTRIES 616 O.OOOE+OO O.OOOE+OO O.OOOE+OO O.OOOE+OO 616. O.OOOE+OO O.OOOE+OO O.OOOE+OO O.OOOE+OO

22 - '

' .. i

'-- ! ' :

' I I '

i I ! ' I

' • I

" I I !

18 '- !

. ' ; ;

16 -. I . I . I . I . I . I . I I .

0 2 4 6 8 10 12 14

(S2z3+S2z4• 16+S2z5•32)/(S2z3 ..... S2z4+S2z5) VS Tz 1 PE

B1 0 . ~ j-iolS/J\B 0

CLOCl!CB 1

MCLOC:K 0

! '-6 ' 0

I I I 6175.2 5992.6 6042.4 6142 6092.2 6191.B 6241.6 6291.4 ,.. ..

IPC Front End Simulation • TO resolution

TO= (sl + 16*s2+32•s3)/{sl +s2+s3)

TO res. vs noise

·TO res .. vs shaping time error

11.04··

11.03· lit

! 1.02 0

';;' 10 a .._ .. g 5 Ill

CD ~ 0

·' ·-' - .

TO R eaolutlon

• • •

1 2 3 Noise (mV)

TO R esolutlon vs S haplng Error (°/o)

• • •

10 20 30 "° S hoping T lme Error ('1.)

IPC Front End Simulation • resolution vs shaping time errors

No electronics noise No time jitter

POI. R•oluttoA VI Stt~g Error ('4)

200·· - • . ~ 150·• ~ g 100·" a: ·- 50 • •

!. 0+-~~--1---·~~~·f--~~~; •

0 20 40 60

S hoping T lme Error (°lo)

Sample 3/4/5 average

Presentation by:

Eric Prebys

Update on Zero Suppr~ and Dynaniic Range Studies for the GEM Liquid Argon

Calorimeter

Eric Pi:ebys, Princeton University

November 4, 1992

Zero Suppression Schemes • Scheme 1- The simplest zercrsuppression. Read out

only those channels which are above a certain threshold. For this sort of suppression, the thresholds must -be kept fairly low to reduce the losses in the low en-ergy tails.

• Scheme !- In this scheme, if a giwn c:b1nnel is above threshold, all neighboring d!annek (fnmt and back) are read out. ~erhaps difficult to implement in practice, particularly at the boundaries between readout modules.

• Scheme 9- This is a fairly simple scheme. ff any channel is above a given threshold, ell d9e dw>nels in its 5 x 5 readout n1odule are read opt, lront and

back. ••••• ••••• ••••• I Ill 11111

• Scle11te 4- 'Thi'5 is sxne•W Yr !k la t gbeme? except tJnat in addition to tAe ~ <d comtaining the aibove-thceshold chamnd, tlpie lpordsjog rows from t!e drrt • NJ! rad,.,

l•I.JrlU~L:..... D••••1;; n•••••"' u~:•-••n

KJ • • • • •n u•••••u rJ1r.r11na'llo

E ff~,~-.·"._·.\~~~ z.,,c

o-t Supp. F"oc.

., ··: .. · .. ' .. .. \, ..... :.~.· ... .. \ ..... ·,_ ' ·.. ... i. ···... ··e ~ ·. .. ' ·.. · .. ' ·. · ..

• Scheme 1

• Scheme 2

'f' Scheme 3

0 Scheme4

... ~... .... ..................... ·· .. .

\ .. ·,. • .. •. "'·,·,.·.. . . ... . . . ..

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

.......... · .. · ..

· .. ~

'········· ....

0.8 1 Threshold (Girl)

Figure 1: Zno suppression factor n. sero suppression threslaold for the varioGs ~ discu.ed.

• 49

•([)/[

<([)/[

.... 0.04

I a.02

.... • .....

<tQ.11 0.1 ,--~~~~~~~~~~~~~~-

1 Cev7

.., 2 ................

·-· ·-· ·-· o-•

·---·-

·-· ·-· ·-· o-•

o(l)ll

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

• . ...

50 GeV7

.. ' --·-······

. -· ·-· ·-· o-•

·-· ·-· ·-· ·-·

-F~ 3: Energy raolutioa vs. llel'O supprssioa factor for 'Ario. suppt"essio. KMmes with (a) 1 GeV; (b) 5 GeV; (c) 10 GeV; wl (4) 58 GeV pitotom.

~12J.'.?ression Scheme 1

";, ______ N_o_N_o_i_se_o_r_P_ile---up _____ ___,(o...,.)--..

0.08 ..

_ NoZaroSupprrrr'or

----- Su~ 11fon Foc:llr •.0%1

•--··----5 10 15 20 25 lO

MiNi119 Et

---· . ' ' Standard Noite ond Pile-up (b)

~ ....... --' '---

.-- No Z.o S...W111lon

Suw111· cm Foc:Ur-.026

MiNing Et

Figure 6: Eft'ect of zero a.,pression on the total missing Er~ (a) without and (b) witla t'--al ucl pileup noise simulated. Tk ~of ~ 1 aero •ppz zwi.>a are ..._,.. by the superimpoeed lmeogrum.

r-------.

SueJ2ression Scheme 2

Ng Ngj• « fflle-ug.

~-- No Z.. 5uppr ••

5 10 15 20 30

(I>) Standard Noise and P"de-~ -

:.:.---1---L~ ... V No z.o ...,,, .. ,,. I r~~-------- Sup pc I ion f'actorw.GJS

• • ··'

00 10. 15 .. ..., ., ..... E'l

Figure 7: Elrect ol aero Aipt111 ·1a • ~ &o&al mming Er spectrum (~) witbo.at ucl (b) witla tMi awl -4 ; ii p llOi9e mnulated. The etrecta el .claeme 2 aero suppr 11 • • 1tt thowa ~ tile s.perimpoeed biatogram1.

Zero Suppression Conclusions

Based on these studies, it is clear that one can achieve a reduction factor of greater than 10 with several zero suppression schemes without greatly sacrificing resolution or linearity. It also appears that the missing Ex resolutioo is not drastically con1pro1nised. Of the schemes studied, it appears that the ~ attractive js the most basic - narnely the scheme in which one simply reads out -channels satisfying a certain threshold. nm scheme al-lows one to achieve reduction factors of up to 50 without greatly impacting the quality of the data. At lower reduction factors. the simple scheme shows similar or superior behavior to the other sC"hemes.

Dynamic Range Studies

• Goal: To determine the dynamic range n~ed by both the electromagnetic and the hadronic calorimeters

• Low End of Dynamic Range: Total thermal and pileup noise /v'21

- ~ 25 Me V for the Electromagnetic Calorimeter

- ~ 140 ~1e \' for the Hadronic Calorimeter

High End of Dynamic Range for EM Calorimeter

Use decays of the form

Choose

M(Z') = 12 TeV

because, assuming "standard" couplings, this gives 10 events in one year at 1<>34s-1cm-2 luminosity.

00 2000 10000 ~

Figure 4: Highest energy in any electromagnetic tower for massive Z' -+ e+ e- decays.

High End of Dynamic Range for Hadronic Calorimeter

Use high-PT two jet events

Choose

PT> 8TeV

because the standard n1odel predicts roughly 100 events -in this range in one year at 1<>34s-1cm-2 luminosity --+

enough to detect discrepancies (eg. compositeness).

0.04 Two Jet Events, p,>STeV

0 0!:-"..1....1.1.L.....•2=0~0~0_.__._~40~0~0 ........... ......u~!!..,,.o.:U:.l.li:.l~LliL... .......... ~,_Joooo

0#/ll Mcndmum ,_en.,,

Figure 5: Highest energy in any hadrooic tower for bigh-Prtwo jet events.

Summary· of Dynamic Range Studies

Electromagnetic Calorimeter

range = ~= x : ~ 288000 ~ 18bits

L "•:>"'"°'.__ ---. .. .a.,J. •• ~ Hadronic Calorimeter

range=~ x: ~ 51400~16bits

Implementation of 18-bit Dynamic Range

• Nonlinear Preampsf

- Easy t9 build, but

- Might be tricky to build and calibrate to desired accuracy.

• Triple Range 1!!-bit ADC?

- Probably available technology,

- Allows 9 bit overlap, but

- Might he tricky to build.

• Dual Range 13-bit ADl"I

- Only two ranges needed, but

- Only 8 bit resolution at overlap.

- Might i1ot exist.

• Some Combination of the Above?

.. . ..... .

• ' ' ' ' .

H~v- M(H)=160 GeV

------ End of Low Ronfe for 121111 ADC.

End of Low Ronfe for 13 bit ADC

~·.,I . : .~ ... ~ \" 0 o~_._...._.._2~0"""-0~ ...... .!:i::::::c400~l..c:::a__.__._eoo.J..,-....... _._...._li.L.D0...._ ........ __._1000J

Figure 6: Highest energy in any electromagnetic tower for H -o 11 decays.

Presentation by:

Sergio Rescia

Sergio Rescia Instrumentation Division Brookhaven National LaboratC1r7

LKr/LAr CALORIMETER FRONT-END ELECTRONICS

GEM COLLABORATION MEETING BNL, 5 Nov. 1992

I B I '

I w 8 I

ENVIR( )NMENT

(10 years \\f operation)

Design of a charge sensitive preamplifier for Liquid Argon preamplifier for cryogenic ionization chamber calorimeters (Liquid Argon, Krypton) at hadron collider experiments (SSC, LHC).

Neutron Flux: 1014 N/cm2

Dose: 30 Mrad

SPECIJ.'tCATIONS Low Noise (trigger sums)

Cn=100 pF-1 nF

tp=15-50 ns

TECH!\, )LOGY

High gm, grnf Ciss' gn/Inss

CnflO<Ciss<Cn/2

W=104 µm (1 cm)

Epitaxial channel JFETs meet all these requirements and are manufactured with a mature, well proven technology. Hybrid preamplifiers using discrete JFETs have been built and successfully operated up to three years. To achieve monolithic integration interdevice isolation must be obtained while preserving the characteristic of discrete JFETs.

BURIED LAYER TECHNOLOGY To achieve interdevice insulation a junction-insulated technology based on a diffused p+ -type buried layer which acts as a back gate has been developed. It uses an epi channel and diffused gate, source and drain to achieve an optimum noise performance.

• Psubs=<>.5 ncm • Pburied=0.002 Ocm • Pepi=<>.5-2 n cm • tepi=4-7 µm • isolation ring doping: >1018 cm-3 at the surface to

prevent surface inversion

N-SUBSTRATE

Im[~] [ v. [ nV] SW,

p FEATURES ld[mAJ SW _n_ sw, ../Hi. oew e ../Hi.. SW,

TEST PADS 1 ADDED 9.2 48.3 0.57 0.49 1.16 200

(ceramic carrier)

TEST PADS 2 ADDED 1.9 22 0.76 0.72 1.05 112

(ceramic carrier)

TEST PADS 3 ADDED 2 23 0.75 0.70 1.07 127

(ceramic carrier)

NOTESTPADS 3 26 4 (T0-78) @ 0.68 1.07 @

TEST PADS 5 ADDED 4 32 0.66 0.60 1.10 136

(ceramic carrier)

NOTESTPADS ~ ,,,,.- ._,,__

6 (T0-78) 5 31 :i 0.61 1.11 ; 200 / =-

too ,.-----,--,--T-~!-t .-..~~,----... . :

·········-~---·-······-r--- ······ ············r···········1··········· IO . . . : : :

:1.:t::l: ::: :t. a)

. . : : : . . ·····:·- -·····-20 .......... ; ............ ~·----------;--·--------·r--·---

l ~ i o io l1,----10~~:---10~4,----:-:1oj,--;-;10~6--;;10'1' --;.;no•

Fig. 2 - Characteristics of the preamplifier labelled 4 in Table I. (a) Open-loop gain-frequency dependence. (b) Spectral density of the noise voltage referred to the preamplifier input as a function of frequency.

JFET RADIATION EFFECTS ( 60Co)

NOISE SPECTRUM

[nV/;IHz] 100....------.,.~,.......,..~-,---,----,-~-,-----,--,--~-,--..,.-.,.-----.,,---.,.......,,__---.,..---,

: : ; : ; : ; ... .. . . . .. . ...... :·· .. . . ........ ···; ........ :-··· .. !·· .................. -~--- .. : .... ····· .... ~- .... ···: ..... :.... . . . . . . . . ...... :· .. . . . ........... ~-· ... .

... ........ ....... ; ..... ············~---··---~------~············· ·····--·~····-~·············+·······~-----~·-··········· ....... : ..... ···········-~·-····

. . . . · · ·· · · · ·· · ·· ~- · ··· · · · ~---· · ·= ·· ·· ··· ·· ··· · ·- · ·· ·· ·r .. · · · 1 .. · · ··· ·· ·· ··:· · ···· · -~ ··· · · r · ·-· · ·· · ·· · · · · · · · · ----·

. . . . . . .. .: ..... ·:· ... ................ -~-- ... ~-- ·- ··········;·· ..... ·:·. -· .. ~-- ·······.... . ·--·· ·:. .. .. . ........... -:- .... .

·-----------~------··r····r············r

.. -~~ . . . . 'f""• ·t···~···········-~·····

0.5"'---=----=-"-'"-~-=---=--=--~=--=--=------'"---"-'"------"---' ......... ~6~"'--' 10 10 [Hz]

.... ....... N

.-102 :c ....... ,.... N .. ::c 1 ...

er ~ .. .. .. ...... er > Ill

c .... ...... >

c: .50• II • 1 ..

~HORT lHANNtL J' FETs • NOi$£'

Noise Spectra Room Temperature

" ' .. ' !i ' 'i i p '

Jij!ii EW

:1: ,j ' ;n ;;;

iii ;;; ;;;

' :p :

. .. - .• 'ttt'

' -,f;! ! ,.

·,;;,,, .

' ' ... . . 4 .. ..;

' j:: ""' 'i 1H-H ' I:: ' I !Ii

--e- NJFET 2500/8

i ! i ij1 · · i' I

il1lli 'J:

. --Jl.JJff0-3500/20 .. • ; !Ji . ! . .,,,, IJJl• " ,. . ' .. H !:l IJ I I! ·:! . 11 ' ; ; · lf ii H

!-Hlf+i+HiHi .Jm'i+'-''-ll-.>-1.ii..liiilli ii rtt-NJFET 2500/20 ·- "'i .+- : , , , ~~---·- ,1; 11;

lo' Frequency

10' 10'

·-- ·-- -- ·- -- .

.._ --.

-• • 0 ,_ >

• . " ' ('>

C:ID ..., .. Q. •• C%1D :::ccn

N cncn Z...I C>:::I .....

• . " a: Q.

z 0 ::c

• ..., z • •

C> C> C> a: a: a: ,_ ,_ , ... .. ~ IN i :::c i :::c I i :::c I a:cn a:U> I a:cn 1' Q.1- Q.1- Q.1-::c :::i ::c => I ::c :::i • zo zcl!lzot,

111p1 rrr r m11111 1111rrrnin111111rrr111p 111rrrrrp·m11111 rrrrrpn' l' rrrrrmp 111 ( . . . . . .-·-

:z :_o

. . . . . . . . . : 0

• • • ;. • • .: •. • ;. • • •; • • .. : • • • ;. • • •: • • • .: • • • : .... : • • •.:. •• • : .. • .: • • •.:. • • • ;, • • .: • .. ; ... •: .. • .: • • • ••• .. ~ID . : : . : : : .. : : : : . :-. . . ·-·-:-

"' " Cll Ol -

••· ---r : z : P..~: :_c~ p . . . .;z . . . . . . . . . . . . . . . . . . . . -... ~-·· ·t .. --~· ... ~ ... ·t ... -~-- .. ;· .. ·;. --~- ···; ···-~- ... !••. ·:'? :· ... ~-. --~- ... ~- .. ·; ... -~-- .. ~- .. ·r-::::

: : : : ·~. : : .. ~:

. . ~ : 7:::::--- : . ~ : : . ~ .

. P"" . . . ':?". •

===--~~ : .

:_z C> . . . . . . . . . . . . . . . . .

-: ... ~- ... ; .... : ... ~-· ··! • ... : .. -~-· .. : .... ; .... !··· .: ... .,:. .. ,;, ... ; ... ;,,,. ·=··· -~· .. !·· .. :_o • • • : ID . . .

:-...::::: .-

..., lC -,_

z .- C>

"' . . . . . . . . . . . . . . . . . . . rill h 1111rrnin11I111rt.·rni1111I11 rf lm1l 1111 Ir rrrf·,·n1h111 lr1·n 1111 h rrrt ffflh 111 r r rrhn·rf111 rhiC:: ~

::c ::c ::c ::c ::c ::c ::c ::c ::c ::c ::c ::c ::c ::c ::c ::c lC ::c ::c lC ' 0 0 0 0 Cll 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Ln o Ln o Ln o Ln o ~ o Ln o LD N o ~ ~ N o ~ ~ N o ~ Ln N ~ ~ N N N N - - - -

LnCllLDOl.DOLn N Ll'I "'- e> N Ll'I " .... - .... -

SUMMARY A monolithic charge sensitive preamplifier has been realized featuring:

• GBW=200MHz • Pdis5=100 mW • Dynamic Range=+2 V • Integral non-linearity=0.2% • en=0.7 nV/-.JHz • No degradation after 100 Mrad (tp=40 ns, T-90 K)

Future Improvements:

• Study of neutron resistance • Realization of monolithic resistors (Si-Cr) • Development of a short channel version •~en- - ~wt ~spo')\..s~ 7~c., ex p~J, Wj~v ~~-

Acknowledments:

Larry A. Rehn, INTERFET Co.

Presentation by:

H. Takai

Precision Timing with Liquid Ionizatioi::i Calorimeters

O. Benacy9;·s, Cannon5 , W, Cleland71I. Ferguson2,.CJ, Finley5,

A. Gordeev6, H. Gordon3, E. Kistenev3, P. Kroon3, M. Leltchouk51

D. Lis~uer3 , H. Ma3, D. Makowiecki3, A. M:aslennikov4, S. McCorkle3, . •.. . • ·• .. • • . 6 . • 4 .. .,..... . '· 9 - . . ' 4 . 5 D. 'Onoprienko , A. Onuchin , Y. Oren , V. Pamn , J. Parsons ,

.· 7 3 ·. ··.3· 3·. ·3 J. Rabel , V. Radeka , L. Rogers 1 D. Rahm ; S. Resc1a , J. Rutherfoord2, M. Seman5; 1'1. Smith3, J, Sondericker III3, R. Steiner1

, .. '3 . '1 . . 3 . . . . . '3 . . 8

D. Stephani ; E. Stern , I: Stumer , H. Taka.i , H. Theinann , Y. Tikhonov4

L Adelphi Universit)'', Garden City, NY 11530 2. Uill\iersity of Arizona, Tucson, AZ 85721 3 . .B.roqkhaven .. National .Laboratory, Upton, NY 11973

~~ ·:---'""':::;!•--,~-:.::o ___ <.,,...:-.~.:. <ll+;! • • - ' -

-4.·~Budket.Institute for Nuclear Physics, Novosibirsk, Russia .·.-·.-=E.t~-P--~~"'-"".-..~:~: 0 ._ :-.. '.:o,:e..-·-·;.: . .'~ :.·.·_ ...... - .... _

· ·, ·, ~;f~IU:Jiib,ia Viiiv~rsicy,-Ne:\Y York; NY 15633 . · · . ; -;~,.--~-J- ~-~:'"-""t - - ·._ ...... - .•.. - . - - ~ - . .

. 6.-oit,1( Ridge National Laboratory, Oalc Ridge, TN 37831 · ·•L ;u(ilversity o(Pittsburgh, :PitulbtiI'gh; PA 15260

s;·suNY at Stony Brook, Stony Brook, NY 11194 9. Tel~Aviv University, Tel"Aviv,.israel

'Precision Ti min 9 wi~ n 1.!qui d Ion iiotion CaLor·1rneters.

• Good t'arncng from Calorimeters:

,.> "> unique assignment of an evel1t to a part\tulor bearn crossing. (bunch ~pacing ... 1sns:

• Conveni"1onql W isdol'V) !

....... _ .. '"~~ ~Scinl'~!~~tQrs ore fast; PMT have sharp ~igool~ ·"·" ·... --·- ~-·~.- · · · 'f.herEifore: "GOOD TlM0

INq" ·-·: . .' . .-.

··•.··HOWEVER'· liquid ioni~otion Calorimeters have.~

~'(" 'f ~u n' orm wave form • very \ow fluctua.t ions l~"\Oti e-/qe\J)

• good -\:im ing is 'Poc;si b le l

Princi P-le of OP-eration (Turko t Sm\\h)

/1 / : _gtt) RC in"\.egtal of flt) I !

Ci' -to , when .f (t) - gt-l) ·15 independent of' amplltude.

Test Setu P-; • Accordion geome~""~ Liquid \<r~rion Calor1me,te.r

• +>beam= 5, 10. 15, 20 GeV

(5x5) TDC stop

• Total Ener9~ __. Flt)= l: a~ .:F tt-ti) 81 t;, chatanet time Offset..

~F .. .. " .. = 0 ....

• ( .. '

, D ~ ' ,., .. E = • .. • ... ' •

> "' +I

I• •

., "' ~ > +

~ ,., ' "' > "'

> • .

• •

e ' •

A . \J\I'•

• • •

" • •

... N • ..l a: •

------·-----

0 .. ' ... "' .. ~ ..

l . ··1· ........... . ..... ··1

.. ·-r-. * : : . . . . . .... i. . .. . . . . . . . . . ..

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

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

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

·-· :~ : ... :O'"":···· ..

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

:> --'tl ........ Ill c

:> ·-'tl ........ Ill c

83 ...... _ IQ--·---··-·--- ··-----c. . c.- ..... . E-- E--

r-:~--~l~-1;-:~:l~-:---------:::~,fl~:----------------------..-C\J

~ -s (!]

-----0

0 .,.... 0 0 .,.... I

(su) v

DRIFT CHAMBER VETO COUNTER

Oo c:n S1

- - - - • .. - • • • -t •I .. - 1-t • - • • • •• - - • • • -[}l CERENKOV #2

CERENKOV #1

HODOSCOPE

51, 52 timed by CFDs

52. ~ Start Countey

82 MUON COUNTER

I - I I I I I

CU ABSORBER

CALORIMETER

On-line Pulse Shape LKr, 2mm stack, 1.5kV Beam Momentum= 20 GeV

Sum of approx. 150 channels

1!k Stopped: 11 o Acquisitions lT I .... '. .... : .... : .... : .... t .... ~ . . . : .... : .... ! ....

.;. .... : .... : .... : .... : .... + .... : .... : .... : .... : .... -~t :1··,: a :..._. "" ~ -::::.. ..

. . . . . . . . . . . . . . . . . . . I . ... ~ . . :-:::=- ................... . j t "!§:. . . J + . "'!Ii=.. . j. .;. • : == -.. . ~-·~···I:···· I····:····:······ ·.·r~~ ·

: ' '; c::.' ;:~: ! : ··· l · : ... [. : ;: . ; [ ; . ; :: ~ 4 . -¥. . I . :

Edge Slope

......... : .... ~t:.J.~ .... r ....................... . . J. i . t 1----. i. • . +

: a : . 1 · ' . t

a " : .... : .... : ... ~~ .1 .. : .... r .... : . -.. : .... : .... : .... i· 1 . + t i . + i{ .;.

.............. !' ... : .... ! ....................... . · · : . + M lOOns Cht J -260lllV SO.OmVQ · · t · · · ·

,,,,;, ... ~.--.~---.] .... t .... i .... ! .... i .... ~ ... . ,........ ......... __ .,....,....., Type

<Edgt>

source coupling 01 DC

lMJ & -2QJmV Holdoff

Multiple trace - variable persiste_nce ! !

LKr 1() 2 c:r-ri;-, ,,,.......,..T"'T'T""T....-.-rT-rrT"T.,..,...,.,..,...,r"TTT'"1~T""T'"T"T"T"TTT"TTT"T"""T"Tl=I . ·'

f /fir\/# rm5 ;: 1.0 ns ·

-5 -4 -3 -2 -1 0 1 2 3 4 5 Uncorrected timing (ns)

780 800

820 S1-S2.

.......

1 1-111.D l.D11

O a I llE100 ·---1.10 ••:I: 1.111 ., .... :I: O.tzt7 4.541 :I: 0.1 ..

6The Correction''

• i. t con be shown that for Flt)· r a.if t-t-tiJ a timing sh\ft 0::. "Lo~~i wi II be lntrodu ced.

• Therefo~e ti co.nbe found offline by mtnimi1in9 events 2. 0 ,_

L €j = L. ( tJ - LOCij ti) jat

toJ. .fo measu"ed -ttme " event j

O(i = Qi , normaliie.d p.h. z:.cu

• This leads to:

/A-IRl =O IA= L. t~ o<lcJ and IR1d =?:. (o(ic.i oe1j)

J•I ./ial

• apply this to al\ runs~ \1rn\~in9 colorime.ter to 3x 3 "b>10er 9eo metr~ .

> a> (.!) LO 0

I 4 I I j B b le I I

> > a> a> (.!) (.!) 0 .....- a ~~

_.,! 141 11 ff \' I

I ~1 t4 I

B e 0 z 0 a> .....- c:

c: «I

.s::. (.)

0 ,... 0 ..... C\I (")

(su) 'l

-------- -

(/) c:

ui 10 -c: :::> 0 ()

--- - ------ -- - - --- - -- ---- - - ---- ---- -----~

+--- a=280± 30ps

-4 -3 -2 -1 0 1 2 3 4

Timing (ns)

- -- --~ -- --------- -- -- -- ---- -- --- "

Timing Energy Dependence

Prohm1nory

_...... ('/l 0.6 i::

OD i:: -- 0.4 s ·-E-i

0.2 1--......--_ - - - - - - - - - - - - - -

0 . 0 L..-L.-1-L-L...J-'--l-Jc...L-1-J.......J-J.......J-J.......J:.....1......J:.....1......J:.....1......J'--'

0.00 0.05 0.10 0.15 0.20

1/E(GeV-1)

6t = (0.183 ± .020) ns ffi (4.16 ± 0.06) GeV · ns/ E

~lta_:.tl): (0./9 :t0-02) vz. 92

Conclusions

• Liquid Ionl iation Colorimeters have excellent t\ m\ n9 pro pert\ es.

• Energy dependence is cal cu lab le.

• the inherent timin9 re.solut\on _..,..,. o.a C:.eV.ns of one tower E

Presentation by:

Paul O'Connor

?a.u \ 0 \ Con no\

Mvo~ t;.o-.-\- tn J

E l.4...-\.o" 1'cl 't'J'lu_.J, ; .., ;j

OUTLINE

Front End ICs for Readout of Cathode Pad/Strip Chambers

P. O'Connor, BNL Instrumentation Division

"' • Performance Specifications ~

• Readout Function Blocks

• Circuit Design and Technology Choices

• BNL Test Chips & Results

• Plans for Future Work

PO'C 1 114/92 lhome2!pocllext.GEM/Nov4.lm

Specifications: /PC Central Tracker

• Pad Capacitance: 10-50 pf

• Most probable Ostrip: -30 fC

• Max. occupancy time: (2 - 4) x 16 nsec

• Desired position interpolation accuracy: 1°/o of pad pitch

• Desired time tagging: within 10 nsec from time of avalanche

PO'C 11/4192 lhome2/pocllext/GEM/Nov4.fm

• •

,. •'

t .•.

' . : t I

. , • I

""'10 11 ol .... ,~ •f 'J ·~ pt, .. u

• • ' • •

' ,. .. ,,,

)1Sc.~1.,.,,.i11.TOk/"'-lr1L l\Jf~t. &

" ,.. ~•li.Crflt. l{ ~,.•ltY ..-o~ \I ..

I f~fi,

' I """'

~'"""" t t>lb 'WA ·. ·-~t

J tE'T

BNLIAU/

Specifications: CSC Front End IC

Strip Capacitance: 8-120 pF

Most probable Qstrip: -30 fC

6 Hit rate per strip: ... 100 Hz (tracks), 50 Hz - 50 kHz (neutrons) r-

Desired position interpolation accuracy: 1°/o of strip pitch

Desired time tagging: within 10 nsec from time of avalanche

PO'C 1114192 /home2/pocl1ex11GEM/Nov4.tm

For Gx Ix• .01, need aclO • .006

aa • (.006) (30 fC) • 0.17 fC • 1100 e·

F.or 1000 e·, Co• 40 pF, CFET +,C51ray .. 15 pF, tm • 25 ns (CR·RC4)

I en.< 0.46 nvfrz I . (Req< 150)

Input device Om> 52 ms (FET), 39 mS (BJT)

le > 1.0 mA (BJT) ... but need to consider parallel noise of ~ 1 o uA 18

Id x W/L > 27A (I) for MOSFET in strong inversion

30% • 5091. from preamp other devices

noise of Rg, Rsub

2nd stage noise

BNL/AW •

c.• '1c.,~ "" • fOON .....

~ f" ' o.a-, ~" IJ .. ,, ("'"") stf'• ~~fl!\~ ( flT)

~· "'( c. 17..A

PO'C 11~111! AlnMA,lnfM'-.Jl•wt.r.FMJNov,..fm

1--" 0 c.:i

For ax /x = .01, need acfO = .006

a0 = (.006) (30 fC) = 0.17 fC = 11 oo e-

For 1000 e-, Co= 40 pF, CFET + Cstray = 15 pF, tm = 25 ns (CR-RC4)

I en < 0.46 nV/ Hz I (Req < 150)

Input device gm> 52 ms (FET), 39 mS (BJT)

le > 1.0 mA (BJT) ... but need to consider parallel noise of - 10 uA 18

Id x W/L > 27A (I) for MOSFET in strong inversion

30% - 50% from preamp other devices

noise of Ag, Rsub

2nd stage noise

8NLIAW

po·c 1114192 lhoma2/pocllaxtiGEM/Nov4.fm

... 100 LM394, J: 2N4124 """ 0 I --.. - 10 2N6483 FET "' ~/2N4250 ... J: ..... > c -... 1 -

~ 250786

0.1 O.lµA lµA 10µA 100µA lmA 10mA

~ collector current 0 ..i:a

"' ... J: ..... > ..s • ..

100.0r 2N3954-8, 2N5196-9, 2N5452-4, 2N5045-7, 2N5545-7

::-,. _ 10kHz

1 t:::::::-=::::-z..:: .... .LWk HZ 2 N 551 5-24, ,~-----:::._--.!./'.'.2,!2N~6~483-5 1 OHz

\2N5515-24, 2N6483-5 10kHz LM394 10kHz

100µA

/(bipolar)

lmA 10mA

MOSFET I

-2N3954-B, 2N5196-9,

2N5452-4, 2N5484-6 lmA

LM394 /c = \ 2N5432-34 lmA J

1 mA (bipolar) , 2SK 14 7

2N6483-5, 2N5515-24 1mA 2SJ72 JmA

0.1 ~---~---~----'-------' 10Hz 100Hz lkHz

frequency lOkHz lOOkHz

• .:!. .(i:

,.... 0 CJ'1

INLJAUI

Charge Sensitive PteamP - Input D . IVtce Ttc/Jnoloqy Cho/Gt

Equivalent Input noise charge:

torBJT,

for MOS,

Conclusions (short tm):

' I I . .... c,.. .1 EN1.- • •1-+•1'•'• '·

1! • 41:T(r +R ... ) '·

.1 I I .-. • 29/1 +0T(i"'+-->

. I' R•i•

2r Kl I• OT(-+R,) +a,'•c: WL e,. l1. ••

I 1. = .UT(R)

!Nit ..

(EQ<I)

J=or BJT, achievable ENC is limited by Rtiti·. p, and fr. For MOS in strong lnwrsion, ENC is constrained by power and ..... allocated to the input ~vice. In practice secondary er.ets like excess noise sourc· es, ease of integration, and accuracy of modeling large input device must be ceNfully considered.

l'O'C-11!1111 ~1/fMIO/lewll()EM/Aufl!4.lfft

•O<J+X

oooc 0001 0

........ :lN3

IPC Front End Electronics

I 1~21: .,.,.m ......

l JS•9 in. e •1 .,.,,..,; ''-2u •• l CIN2l00I..

-.&"'"

preamp_n ?/"'18/0'l ~ I .JL

"' u .., I u u

.. .. u

•• l8eu

~' N . - ~ • l

f • j

• ' J,

Clb"'""" . l•Zu

i··219u I '

• vb2

PO'C2/26/W. -1/1

~ 0 OQ

Shaoer Design

• Noise (power and area tradeoff)

• Peaking time control

• Dynamic range

• DC and AC stability

BNLIAUI

PO'C 1114192 /home2/pocllext/GEM/Nov4.lm

'"""' 0

iNUAUI

CMOS Lowpass Sectlqn tac $fmlq1uglan_Shfp«

'to• 2gm31C1 - (ibias)112

2 C1 P5 8m1R1 ~P5 C; R1 ·

Q • 8,,.3JJ) (R1 +R2)CL = 4JJ3JJ3CL (RI +R1) •>

• pe•klng time depend• only on (ibl••)

• •h•pe of pulM le d1termlned by component retlps, lftd9Dendent of (Ill•).

f'O'() 1123111! . "'91M1i,ecAelll!GEMIA .. 24.lm

.... ,_.. 0

1991-2 Prototypina Actlvltv

• 9 MOSIS Tiny Chips submitted

• All 2um CMOS I BICMOS process (Orbit Semi.)

• Investigated: 1000012 NMOS, PMOS test transistors

Preamps:

Charge sensitive folded cascode (NMOS, PMOS input device)

45 kOhm transimpeclance (NMOS, bipolar npn input device)

Shapers:

20, 40 ·100, 500 ·1000 nsec unipolar CR·(RC)4

Output Drivers: source followers

Preamp-shaper, preamp-shaper-tracl</hold with output drivers

• R~O·Ma~ r~" C MNlf.\~ A"'-~'"" •~_p•' ._ •~ '/~1

PO'C 1 T/4/f.:I 11..---""---•-·-·-.-··-·---" •-

I-" I-" I-"

• Components

R, C, MOS, BJT

noise vs. frequency, bias

radiation

• Circuits

Noise vs. tm, Cd, bias

Linearity, dynamic range

Peak time

frequency response

Loading effects

Crosstalk

BNL/AW

PO'C 1114192 lhome2/poc~ext/GEM/Nov4.fm

C! -"' 0

I h ...............

..........

.... ....

-·~ ... ...

......... . ..... ..... ... ,.._

..... .... -

ICS MOSFET Noise vs. Frequency -- Id - 4mA

If~ -~ ' ' ..... ............._ -

' ....... :-..... ,, r-. ' ...,.., ' '

........... .... ~ ~ ::::: . ......._ ...,..

-I -... ,..... -

' -- .... ---- I . ,..

... ' ...

100 1000 10000 100000

°r' iL\.,.S

\'.' ... r

; . I >.lo - l '• r r,,r . ..., e-

112 - ' ,. - •10 -1 $' :r PitiC

-• c ... c E • a .c • u it

. ' I ' •

!l o~ "2 §2 < .... §Ill ~~ ;~ <N ~N . . . . • .

• • . i1 . . . ' . • • . . . .

• 0 0 I I 'i' J . .. . . • . . . .. ' • . • . •

/ • ' . •

• • • • • •

. . . . . . ="-·• --• • • • • • • • .·/-. .

• • • • • • . • • ' • • • • • ' . • • • .. ,-.. ' • • •

~ • • •

_;.....

• • • • • • • •

! E• i;iji

~r i j

- - - - --

c: d E ca ~ a..

LO (,)

..... .E Q)

E i-0> c: ~ ca Q) a..

• VJ >

(,) z w

: •• ; ..• : .•• ~ •..• ; ..... ; ....• ;: •..•...• : .•.•...••.. ; . . • • . . . • • . ..••. ·-·~· '1(;'4-!el· ... : ..•. ; .•.. \ •..••. ;. .• : .. : ... ;. .. -: .... :- .... : ..... i ....... -: ........... :- ......... . . . . . . . . . . : .. : .. ·:· .. ~ ... ·:· .... , ............. ~ ........... ~ ......... . . . . . . . ... . . .. . . . ... . . . . .. . . .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...................................... : ........... ~ .................. ·:·. ·: . . . . . . . . . . . . . ·:· .. , ... •'• ........... ~ ....... ~ ........... ~ ......... . ' .. ' ..... ········:···········~··················-:···:· . ' ..... .

~ .. : ..... . ...... -: ........... : . . . . . . . . . . . ...... ·.· . ·.·

O<l+X<> . . ............. -...... ~ ........................ .

: .. ; .. ·:· .. ~ ... ·:· .... :- ..... ; ....... ·: ........... :- . . . . . . . . . ....... ·:· . . . . : !··!·· ·:- .. ~- .. ·:··· ··:···. ·!· ...... ·:· ......... ·:·· .............. ··:-· ... ·:. ; .. ; .. ·:· .. ~ ... -:- .... :- ..... ; ....... -: ........... :- . . . . . . . . . ........ ·:· . . . . . . . . . ...................................... :···········:··················:·· ; .. ; ... : ... ~ .... : ..... ~ ..... ; ....... ~ ........... : .................. ·:· ., ... , . . . . . . . ······-········-·················· .. ···········:····· . . . . . . . . . . . . . ... . . . . . . ... . . . . : ..... : ....... : ........... ~ .... . . . . . . . . . . . . . ·'· .. ~ ......................... •' ........... \ .. . ' ' •'• • • \ ' • ''. • • • I • • • • . . . . . . . . . . . . .

; .. ; ... ;. .. ~ ... .; ..... ~ ..... ; ........ ; ........... ~ ................ .

~ .. : ....... ~ ... ·:· .... ~ ..... ~ ....... ·: ........ . . .......... " ......................... " ........ . . . .

····:··· . .... · ... . ····,· ..

.........

..........

. ....... .

: .. : .. ·:· .. ~ ... -:· .... :· .... ...•.. ·:· . ·:· .. :- .. <· ... : .... : ..... ·:· ..

0 0 0 0 -8 0 LO

0 0 0 -0 0 LO

0 • • • T'"" ........................................................................................................... . . . . . . ... . . . . . . ... . . . . .. . . . . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . ... . ... . . . . . ... . . . . . . . . . . . . . . ... . .

0000~ ooos 000~ oos -a ·s·w·J 'ON3

oJ ~· c:l • -· E! 1-1

§ • • ~

• " • -I ~

.. ~·

'\ \ • \ ~ ~

"' ... , "·" .. \

.... \ . \ .

\t \\ .. .. \

\ \ .. \

ll, I a;; •• !1 b I

••

• \ \

\ '-1-----~---->o-"" • I I I I • ..........

n I •• is Ji I

· .. I I I . I •

..... -:.a

I I I I • .~ .....

.... ;1

• " I I

• " ! I

• " ! I ,.

.--...-,..--~~~~~~, I \ ·. & ...

\ \ \ ..

\ ... \ ·. , ... , ..

'" '" '· \ '·

h :: . •• .. -· Iii •1 f. I

• ·~ ~ " ... ,

! I ..........

... .. \ ... \ .. \

• -,__,,..~.-------. I

\ ' \ •

\. "' ·~ '.,

\, -..~

"' ..., ~It .. \ \\

I ..........

! ._.,,.. 118

~· " \ " \ \ \ \ \

" \ " . I

.... ;1

I .. I

; .. I

.------.....--------. I

I I I I • ·~"»cl .

I ! I I • ..........

• • ii: I

• • .. ~

I I " .

R \ ' •' I \ :2 ..

\ !; \ !1

~ .. \ ~ ~ \

• I : I ~

~\ I: • " ,:

• I I I • ,_.,...

•, \ \

\\ \ \

I I ..... .,":lfrr,;]

I I ..... .,')NJ

\ : ,: . :

• •

• • ~ I

• • g ~

\ \ .,

' \ . .. . ",, \,, ~ :; .. ", \, ;.: i!

\ . il \ ~

\, \ \" \

I I ....... ,':>NJ

\ : \ : \: .:

I I •

\,: \ ',,a \

',, \. o\

\"\ I I !

..... .,-i-. ..

.. I • ~.

• • N

• • I

• \ \ \

\ \ ..

\ •, \

\ \ \ •,

' \ \ I \ .

'• \ \

"' \ \

\ !,11 . . . -· : ::.: ... : ="': : ~! lil

\~ ~ II!~ : ~ .. . .. \~ : il

\" \ \ .

\ \ : _,

i i ! · . i. '! •1! : i i i ii i i i i i i i ii i ! i i i i i : i i i i i : i I ·rr:····-1.--.r···:··;·:·:r:r··········:···-·:····:··:··r·r:::··········r··-··1··-:-··r··t:·:·::··········:·-··:··:H:··r:·:· ::: .: ii!!:::: i :::::::: : :::::::: : :::::: I : ! ! : : : : : : ·:·: : : : : : : : : : : : : : : : : : : : ·: : : : : i i : : . i : 'i i ; : : i: : i : : : : i ; : : : : : : : : : : : i : : =· : :

•f"f"1•--••-J•••••'"t•••+••f"•f"•f•f""tf-• .. •••••••f••••••f•••-J••'1'•"t•f"-t'"t't'••u-••••'f"•••••f-••• .. ••f••t•"t•t•t"t••••••••••~•••••~•••'t'•-f-•,.•t•t• :.: : : : : : : : : : : : ·: i : i : ! : : i : : : : : : : : : : : : : : : :::· t ::::::-:: : • .:.::::::: : :::::::: : :::::: : : : ,. : .: : : : : ': : =~~ : : : : : : : : : : : : : : : : : : : : : : : : : : : : I i : : : : : ~..c, : : : : : : : : : : : : : : : : ! : : : ! : : ·fat-t-··----+---.. ···1'··-r--t·t·++t-·-···- ·······t··-t·1'·+-t-t+··········+·····t-·-t-··t··t·+·t·t~·-·······"'?··-·t-·-'t-'··t-·t--t·1· .• ·-· • 1 •.•• ·-·· • • • • • • •• • • • • • • • •• • •••••• j : : •. . ,.. : .. :. i : ! i : : . : : : : : : : : : : : : : : : : : ! : : . : : : : : :2: : :-:•:.:: : : : :::::: : : : :::!:: : ~: :::· OOo • I ••• C11•• t I I ltOltl t I I 10100~ O 1 o 100:

: : ; : : : I I -: ;. : : : : : : : : : : : : : : : : : : : : : : . : : : : : ·ttr·········1······t ···t··t·t·~·ttt··········t ... ···· ! .... ·t··1 ··t· t ·t·t t ·· ········t-···t··· t·· r ·t·t· ~-! -~ · · ······· ·:··· · · ~-···t·:·t 1 ·t· ::: : : : : :::::~: : ~•ci.d= : : : : ::::: : : : : ::: i ii . ~ i ~ i i i i ~ A i E ~ .. bu: i ~-~ i i ~ ~ ~ I~ ~ ~ ~ ~ ~ ~ i

.f.f.i•••••••••••i••••••••-+••+•+•i•++••••••H••••i••••••i••••i•• .. ••••+••++••••••••••+-•••••i••••i-••i••i•.)•i•i•i-•••e.·••••+•••••+••••••••·•·•·•• ::: : :::::::: : :::::::: : :::::::: : :::::: O O O O 0 I I I t I 0 0 I I I I 0 0 0 0 0 t O I O O O O O O O O 0 o t t o . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ::: : :::::::: : ::::M.ci: : :::::::: : :::::: l il l !P!!..L : ;:~· cotJ l ! ! l ._, . : ! ; ; ; ; ! ! l ; ! ! ! ! ! ! ! ; .,_ +-r·· ........................ .. r.: ~~!& ~-4 .................. •·f··•·I ..... w;., ·Wr.ot •••••••• ··+·····4· ···I·•· I ··I·+· f·I ·) •• ••••••• ·+·····+···¥-+·+-4 ·f· • • • • :,op• 'Iii' • • ':'£<.~ • • •• -ACC. • • • • • • • • • • . • . . .. ::: : :.':!!:!: .·:::::::: : :::::::: : :::::: .. • • .... ,Dllj • .. .. • ~ • • • .. .. • _.,....,. • • • .. .. • • • • • .. • ; : : : : 1$t': : : : : : . ~ : : : : : i: .. : : : : : : : : : : : : : : : :

-i-i-i---········i-·-···i---i-·-+ .. .i, .. i . .i,.;.f....... . ~ ........... i ... 4.4.+-i-1+-------~-i----i---i--i-++• .. ~---·······+·---+--i-+-i.-i-i-: : : : : : : : : : : : : : : : : : : : •*•' nr : : : : : : : : : : : : : r ii i· i i 1 i i i - '"'°·' : i i i i i ii : \ : : ; ~ •. I~ ~ i i ; i ~ i ... . ..... . :. - . . . . ... ·~· . . .. . ..... . . Ll.S ............. i ...... J_..l.l.l.l.ili .. ~~-: ...•...... i .... · : . .i .. !ll! ........ - ... i"1:.·r--t .. i....i.l..i .. !l ........... i ..... L-.. l .. L .. l.i.i. : : : : : : : : : : : : \~;:.. · .. : : : : : : : ._..,-. : : : ; : : : ; : : : : : : : : : • : : : : : : : : : I .,,.i. : L' 0 : : : : : : : : : : : : : : : : : : : . . . . . . . . . . . . . ' . . . . . . . . . . . . . . . . . . .... ::: : ::;.:::;: : :::::.·= : :::::::: : :::::: li-i----···--i·---..i .... -l .... l.J. .. j .. J..i.:.. ............ i ...... i ...... : ... .:. . ..:. . .:..l ........ l ...... l .. -i--:....; . .:..: .. w. ............ .;. ....... .;. ... ..:,_i,. .. i-..l-i-• • • • • • • • • • • • • • • • • • • • • • • • • • • • t • • • • •• :: : : : : : :: : :: , ..... D_,; : : : : : • : • : : : :: • : : : : : : ::: : :::::::: ••.---s::::::: ··::· : :::::: ! : : : : i : : : : : i HiftAui : : : : : : : : : : : : : : o 0 o t I o t o I o o o t 0 0 o o o o o o o o o o o 0 o o o . :·:-:-·-··--.. ··:······? .... -: ··+·+ -:-++ fo··········:·· .... :·· .. i··-: ··:· -:--:-:+. --······+-........ !·· .. :··. : ... :-+· :-:-:·-···.. .. . ... : . ::: : :::::::: : :::::::: : :::::::: : .:: .. . 0 0 0 t t o • 0 0 0 0 a 0 0 0 o o o • o o I • 0 o a 0 o o o 0 o o o o o o

::= = :::::=== = ==:===== : :======= ·c· ... ===== 0 0 0 t t o 0 o o 0 o a t. 0 t t o o • I o 0 o I 0 0 o o 0 o 0 I 0 0 o o

' f.in••ss11..;.1ss1siM11f1ai1iiisi111iriiii1111s11t111i111s11i1111i11.;1.;1.ai.fiuun11ssii-s1ts1i1t11i111i11i1.;.1;1;.;,, .. ,.... "'''-••••ti11i11i1i1 o o o O t I 0 ·1 I 0 O • 0 I O t 0 o • e 0 o 0 0 0 0 a O O O 0 0 I I o o o ::: : :::::::: : :::::::: : :::::::: -: : :::

t to too tooo Jftr ,, Tm,llMC

~ · 1

: (v(i")-V:(~~;)- fast & Slow Response to Delta and 6~/(;:cY:)~~(;:,~ IC

f-" _20 V: (V(12)-VOC(12))

04/30192 18:01:46

l'J.0 1. •

-.40fl I " I\ '-: X I k>S I

- .tlfl!

-.11e1 ' I ""' I r I I ..... --i.eL------t---...:s~~-t-:-7~_;__---t-----:~ -1.2r I I I I I I I I I I I '· I I I I I I I I I I I I I I I I I -, I I I I I I I I I lx11-6

8 - : : : i , : ~1! : i : : i i : : : : i : ! : : : : i : : : : : : i : i i : : : •':•!•!•••••••••••:••it•:•••"f'••.,.• "t"•!• 'f"~!•• ••••n •• •f •• ••••f••ut 0 "f'•""t• 'f" "t "f' 'f"••• ••• ••••'f" •• ._,.!_. •• !•••f••f•"f'• t•f't • • • •• •• u • .,. ••••• 'f" n•!•.. •• !"•!" •!• •: ... • • : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . . . . . . ..... ' . . . . . . . . . . . . ... . 0 0 0 0 0 0 0 0 0 0 0 t t t t t 0 0 I 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 . . . . . . . . . . . . . ' . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ' . . 0 o 0 0 0 I I I 0 I 0 0 I 0 • 0 0 0 I I t t 0 o I o I o o 0 I 0 o o o 0 o o t 0 0 I t 0 0 0 0 I 0 0 I 0 0 0 t 0 0 0 0 0 I I 0 I t I I I 0 o I I 0 0 o

· :·:· ! ............... :······: ···:· ·:·:· :·:::·· ···· ·····: ···· ··: · ·· ·: ·· ·: ··:·: ·:·::-· ···· ···-: .... ·:-.. ·: · .. :··: ·:· r· r: ··· ·· ·· ·· ·:··· ··: · ·· :-·· ·· :-·: ·:- ·:·-·· · I I I I I t I O O O O I 0 f I 0 0 t t tO t t o o o t O O. f O O I 0 ' • IO t o O o I o o o Io 0 ' o 0 I I 0 0 0 ' I o o o O o o 0 o I O O o O I ::: : :::::::: :~:t::::: : :::::::: : :: :::: • • • • • • • • • • • • .!16\1'( • • • • • • • • • • • • • • • • • • • • • • • • ::: : : : : : : : :: ®' k1~" : : : : ::: : : : : : : ::: : : : : : : :

·f·f·l···········t······t····t··+·+·t·++t····oaoooO ,.M: •• ,.i ••• ,. •• -t··t·+·t·t+··········t••••ootooootoooto•f•-t·t·t·~-----·····-t·••oo+oootHO --1--t·t· •toOOOO t It I I 0 t t 0 I I I I I O O O O t I I I I 0 I 0 0 o 0 I I 0 I 0 0 • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ' . . . ' . . . . . . . . . . . . . . . . . . . .... '. . . . . ' . . . . . . . . . . . . . . . . .... I 0 f O I 0 0 f o I 0 I 0 0 I I I I I 0 t I I 0 t 0 I 0 o 0 0 t O I I 0 I . . . . . . . ' . . . . . . . . . . . . . . . . . . . . . . . . . .... t ff 0 f I 0 0 0 0 I I I 0 t 0 t I 0 f 0 f f t I 0 0 I I 0 I f 0 I 0' t . . . . . . . . . . . . . . . . . . . . . . . . ... '.' . . . . ...

-i·i·i·-·········•····--•···+··+·+-i-++i--·········i------i···-i---t--t-+-t-t+---·····--1------1----i··-i--i-+-i-i-i-··········-=-····-+---i--- --1--i-i- -i····· : : : : i : : : : : : : .., ___ ..,,: i :~· : : .=ltd::: : : : : : : : : : : : : : : : : ..... ' ......... ~-· '. . . . . ...... . ..... . ;;; ; ; ; ; ;;;:"i"t. ; ; • •n;.Hd OQl)gi,nd; ;;;;; ; ; ; ;;; ; 011 0 I t I' I I 00 f I I • .... w ..... •• 1 o 1 O O O IOI O O O 'o I '

.f .. f.f ..................... j. •••• j. ... ~ •• .;. • .;.~.~-f----····--·i·-·--·i----i---···•·<··>·>.;. ........... ~ ....•. f •••• i---i--i-->-i-i-;. •.•••••.•. J,. ••••• ~ •••• .; ••• --~·-···· ••••••• . . . . . . . . . . . . . . . ..... ' . . . ' . . . . . . . . .... : : : : : : i : : : : : i : : : : : : : : : : : i i : : : : : : : : : : : : : : : : : : : : : : : : : : : : u ... ; : : : : : : : : : : : : : : : : : : : : : : • : : : : !:l!c.ctJ : : : : . ........: : : : : : : : : : : : : : : : : :

-i- i- i. --........ } ... ---·~~-~ q .. + -i-t>l· •·· ---·. ---.J.:···· · i -·· · J •• -i -~f. -W ...... --.. { ...... i--.. i ---i. -i -->· i. i-~ -· ..... -.. .:. ..... '-'·-.{... .. i-.j. i. -i ..••. : : : : :/6D; 9 : ; , : ~~'-'I: : ; : : ~Mt.c, ; : ; : : : : : : : : : : : : : : : : : : _. ~ : : : : : : i : : : : ! : : : .. : ! : : : : ! : : : : : : : : · · . . .... ~~ · · . . . ..,.Ill>·~ · · · · · · · · -Ph"" . . . . . . . . . . . . . . . i ! ! i i SS f : : i i : ! :\c..oN: i ! ! : i ii tP" ! : ! i i i : l i ! i i i i l i

·i·~·!·····--····t·--···t····r··t·t-~·tti--··· ·•---···!·--·!···r··t·i··r·tt·------···i-4b-,i····!·--?·-!·t·!·!·~---·-·····t·····i·--·i··· --~·-:·~- ·!·-··· : : : : : : : : : : : : : : : : : : : : <P.u.c:.~ ...... :- : : : : : : : : : : : : : ; ;; ; ; ; ; ; ; : ill 1=• : ; ; ; : ; ;; : \ ; : ; ; ~ f>f"'l.L.. : ; ; ; ; ; ; : : : : : : : : : : : : .. ~-:- : : : : : : : =~ilL).f ! : ! : : : : : : : ! ! :

·rr~---··--····~·-----~---·r··r·1· 1·r 11· · ·· ··--· · ,. '::} ~ · i : --~ -r ·r1 r···· ···· ··r ~ 4-~--~---t· -~ · 1 · t· t1 ··--- -----1---· -r--·r· · ·- ~- -r· ~- · ~ · · ---... . . . ' . . . . . . . . . . . '. . . ' . . . . . . . . . .... : : : : : : : : : : : : : : : : : : .. : : : : : : : : : : : : : : : : : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... . ,:.,:.: ........... ; ....... :, ..... ..;. .. ..;. . ..;..: . ..;...;.: ............ ;~ ..... : .... :. .... :. .. :. . .:..:.-:...... . .... :-..... ; .... ; ... ; .. ; . ..;..;.;.:, .......... ..;. ..... ..;. ... ; ... .. : .. ;.; . . ; .... .

i : : : : : : : : : ; : ..io-1.l : : : i : : ! : ! i : : ! ! : ! ! : ! : : ! : : : : : : : : : : : : : ®''""" . : : : : : : : •. : : : : : : : : : : : : : : : : : : : : : : : LAi.a\ti : : : : : : : . : : : : : : : : : : : : : : : : : : : n:w I : : : : : : : 0 • : : : : : : • : . .:.; ..... ------:.--~---:.---4"· -~- ..:.-:-.:.~.:--......... ; ....... ; .... ; --4. .. :. • .:. ... -~..:. •• --·· .... .;.. .. .. • • ~: 1, .......... ..;..·· . ···--=· ... ":· • • . :. -~ .... . : : : : : : : : : : : : : : : : : : : : : : ... : : : : : : : : : : : : : : : : : : : : : : . : : : : : : : : : : : : : : : : : : : : : : '· '· - :,i.Lll'li, "' '· . . • ' • • . . . • • • • . • . • • • .. . v ••.• . ' . . . . . . . . . . . . . . . .. ' . . . . . . .

1i1i1i11111uuuj11unj111ei11ii-1ii1i1~iiiu111111111i1111ui1111i111juj1~1j1j.111u11111C.11111 1111 111f:11 ••• 1i1i1u1111111iit11111C.1nii111 n!,"~:•!,•!,•!:11111 ::: : : ::::::: : : ::::::: : : :: : : :

1 10 100 1000 10000

Tm.nsec

~ j> ..... ' ,

I I f ...

i uf' ~w

t -• ... ; Q : .s J I f c !! ~

• .... ... ... ... ... 0 ...

• ;; ... Q

! .. .. ~ • ~ !! • I

.. I l!' :ii : ~

I! •

001 • • .. .. 0

• 124

••

0 • -0

-• !'> 0

-.. .. 0

0.0 0.2

• • • • • • •

• • • • • • • •

~ 1si$:ials • 1ooymax(voutp)

-4-4-4·2 0 2 4~ .

=-....;

°' -0

0.0 0.2 0.4 0.6 0.8 1.0

(~. 100),tm&Jt(VO&Al'I)

·3 ·2 ·1 0 1 % )

Presentation by:

D. Crosetto

Algorithms examples

Local maximum

I C > Ii for

Threshold

i = 1, ... , 8. 8

I I I < ~ Ii + C. 1=1

EM~~ IF c >-

---- ~o: EM~>- EM~

@:lil..---' c 1

(BAD) Threshold < c .. + 1 .. ; (EM) < Threshold

THEN "' x "' almpllflecl trigger tower

Check for ISOLATION

Threshold <

Jet finding

EM:> HAD~~~~

~ .. ~ c. + L; Di+ L; ~+ L; 01111t

1=1 1=1 1=1

Threshold < ~ EDft + 2; Eat 1=1 l=l

\ I ·. /

-- I'll s:: ' ~ Q) I

-i Q,) 'ti I ..., s:: I'll Gii I

"""' ' Q) I

~I ~ •• ... 0 ~ -Gii

Algorithm flow-chart

Phase 1:

Phase 2:

Phase 3:

Corpus Christi 10/2/92

Calor1iwtry In HEP

aet code

aet code aet code

O'lltp1at oodo ~~~~~~~~~~.....__~~

1 two • -· - (aorth 1 z a) > tlarwbo14 a two ·-· nm (out 2 z 1) > u.n.ut• .. "had" r-· (north l z 2) < U..-b•l•

• ·1i..r r-· <out 2 z 1> < u.rouo1• 11 laolaUon aoblnocl

S2 p.-1110 Jot fo'lllld

e.1. A •np" may return a code = 37 (1+4+32) st.a.Uni:

a possible electron wu found, 13 0 - but it wu not isolated from the sorroundine ener17,

- and that cell may be part of a 4 x 4 jet.

V./ e si

3D~Flow Cell Architecture

Top ' 2

HAND SHAkE

,, .....

Bottom

Dario Cruullo

SeptemMr Z4, lff2

I , .. I I~ I I~

lo , .. l UJ

I ili Jl

I~ I ro

I~ 1 ..

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

"""" ~ N

--------·-·· -·--··--· -- . --- . --

EP assembler code of the four pipelined stages of the 15 steps algorithm

icec tep

• ' I • 7

• • 18

• • 10 u

common code to all stace•

.oell 0,1,2 loop&

rl••-"rH rt•-t"rll r7•rl+a. .. ...... rt-ra+e, rlO•d+., du•r7-rll,

~· •ortla. du•rl-rll, ......... ltna-u ... ...,.... . ....... -.... ... ,._ ...

lO , __ ltn•-a. ........... u •• II u •• 11 ... •• II H ti u 11 ti H ..

--= ..... : -•1: ·-·= •-a~ ....

-••oe-rlO, ..... -... 'bn DOHD41, .... .... aop, lt•Oll .. ...., .......... .... .... ,. Prt,ltnloop .... aop, aop, .... ..............

Different code for different

Staie 4 I Staie 3 I Stace 2

(•} (A)

Its\ (le) Its\ (14)

lt•t (le) Its\ (14)

lt9\ (le)

••t (le) ltlO\ (14) ltlO\ (le) ... , (14)

ltlO\ (14)

(•) ('4)

(M) ('4)

.. t (14)

(M) ('4)

(la) (llt)

ltlO\ (•) ••t (41t) lt9\ (lo) lt-t (14)

.. , (lo)

ltat (lo) 118\ (14) ltm\ (lo) •• , (14)

118\ (U}

(le) (14)

(lo} (14)

ltm\ (U}

(lo} (14)

(la) (Ill)

ltlO\ (la) .. , (llt)

.... , <••) lt•t <••) lt"'t (lo) 11•\ (••) lt"'t (•II) 11..i (lo) II .. \ (14) 11•\ (lo) lt-t (14)

lt-t (14)

(lo) (18)

(le) (H)

lt•t (1•)

(lo} (Id)

staces

Staie 1

(la) (lit)

... , (la)

.... , (tit)

11•\ (la) lt-t (llt)

lt•t (le) a.ct (la.)

.. , (•) ltlO\ (•It}

118\ (•) 11•t (A)

(lo) (14)

lt•t ('II} (lo) (1•)

.,., (u) ltm\ (••>

(lo) (1•)

Comments

......... ·-· ...... ,.... ealorlm

........ "laal' 'lal•• tnm ealorlmeter aortll l z I ·-· -aortll l z I "laal' nm ..... , ......... eut I z 1 "laal' -oom,.... l z I ·-· nm to Tlanolaold

I oompue I z I ·-· nm to Tlanelaold

I ·-· • ,.,.._., ... (l s I) I "m • Tlanelael4 - "lael'

: "-· • Tlanelaol4 (I s l) ; ·-· ............. - .....

I .... o•t Wwv 14 .... oat 1 z a•.,.• ea.era I HD4 ••t tower 14

: -4 oat Z • 1 ·-· ••ra

I_. - a.U ftlee I -4 .. , a.U •alee

line code

• • I

• ., • • lO ll 11 ll H 11 ll 17 11 11 10 11 22 23 H 21 18 l'1 21 H 10 31 II

.l.~~~~.L-~~~-'-~~~__J~-a = inputted ·em· Talue -·---- ·-c • outputted tower id b ::s inputted •bad" Talue; d "" output\ecl •em" sum (either 1 x 2 or 2 x 1)

··.•.· 11111111~ 1~!ll7::~a~

.. i.i l l l l l l l I\ ·.:.~:;,:'-\;_·,· ..

~~;.;,::;:.\f,::<~:(~:;l·'<~iit.;\:::,·:',~::~,~~,Y;-:· · ...... -.:·_ .. , . . : . '": -,, .:·:·:':·: .. ::-=:-:: .

..... "" Cit

'ta1e St

Pipelined Parallel Processing Architecture

/' u ft 1e 3/ •/'

Stage 2 Sta~e 1l'

....... , .. .............. ~~l•/I/ ..

--. • • • • • I • • • • • •

..... c;.,

Timing Diagram of Four FEP Stages of

a Pipelined Programmable Level 1 Trigger

- electrons - jets ~ -

Latency

!st ... •I Is.._. 11 lst•e• 2] lstae• 11 E Raw data . . l - .i. 0 ns +_l__L

Algorithm I ~~~/~=-~ execution i;;; time -----_,.

.........

•••

112 ns

128 ns

144 ns

160 ns l2 ,... nJQ II 1- I

~ IWll l+11Df176ns

..,.. ~ • " 11 •" I I . w 1 9 2 n s

I __ ___ . ·------- ___ _ _ ... _ T_1µi_e Corpus Christi 10/2/92 CQlorlMetry In HEP DQrlo Crosetto

Question asked at this point:

• how can you control the data-flow interruption in a thousand

processor array ???

Answer:

• watchdog circuitty at each 3D-Flow processor

1' j "" "1380 1' 3AU:°31A39U2 9U 1' ·----~~--- - - - - - -

One Channel of the'' :3D-Flow''

Pipelined Parallel Processing Architecture

Staae 8 Staie 7 stage 6 Stage 5 Stage 4 Stage 3 Stage 2 St.age 1

.-... ~ .-... ·~~ 1 ~· w; ~; 1 !flt 1 ~f~

Triccer = Watchdol \

Watchdog Data-Flow Control SSCJ. 10/23/1992 Dario Croaetto __.

, .. . I

I~ 1: I~ I~ I (\ It

... .J d"C!> Vl::JOV • °Jf IV...Jlf I~ ,.....,<_., 1\J • --------------------------

.....---------··· I I

One Channel of the ":3D-Flow" i I

I Pipelined Parallel Processing Architecture

I _.. ~ :.o .• , .,

Stage 8 StagC' 7 • • • • • • • • Stage 1

~4 '' • u ,'\!_. ~ ·' '7.

• .r

'"""" •• ~., ... cl'--~

1 ' , .. . I

: . I I 1...1 I q.

' I I~ I I I :s: i ' UJ

I I~ ! , ..

I ~ I I !ii I I~

c I ro

... 0 !l

@ .. rt-~ )"\:F i'\l\ I ; ~ '-B@~:

& & ,i;:g:~. : :---=~--J__----------~--L--1

___ -··. -········ ····----- Co~trol ~=j . Watchdog Data-Flow I ll8CL 10/U/ltn ._ __

.. ..JV''-" "'-'UV 111· ..JI 1v_,., 1J 1vv .,,.,., I -~4- - - - - - - - - - - -One Channel of the "3D-Flow" Pipelined Parallel Processing Architecture

1 1• I " 13

I ' i-~, ~i~ 0

~~fl l~~fl

Sta~c 7 • • • • • • • •

~ .1'1.:;:1JS~~ "' . . . •••

k~t ~i'

Stage 1

I ...._~~~~---~~~~~~~~~~---'

11iuw • Wateb... :

Watchdog Data-Flow Control ac:r. 10/11/1198 __ _ _______ .. __ ·- ·--- _ . __ ---· -·- Dario Cr..u. 1

I :s I~ .. I~ I c ,~

Dart• c ........

---------·-·--·--· ·--·-

2.060!.016· SQ -A-

!·. 9001

1.2!0:.010 SQ

1. 60:.008 SO

.025R, .015R IL.Xi IL.XI

I !.X I .Of.OR

NOTES: I. GOLD PLATE 60,.---22s,. INCHES OVER so,.~3so,. INCHES

MIN NICKEL. 2. HEAT SINK TO BE ELECTRICALLY ISOLATED FROM ALL PINS

3. SEAL RING TO BE ELECTRICALLY CONNECTED TO VSS. 4. MATERIALS SHALL CONFORM TO LSI SPEC J800-000002-00

LATEST REVISION AT TIME OF PURCHANSE. . 5. ALL VIAS TO BE SOLID SUCH THAT VOIDING IN VJA

WILL NOT REDUCE THE VIA CROSS SECTION BY GREATER THAN 20 /.VIA DIAMETER.

10/76 VIA 3 LTPDI

6. T-BAR STAND OFFS TO POINTTOWAROS CAVITY CENTER.

7. LEAD RESISTANCE: SIGNAL---10 MAX. B. REFERENCE VENDOR NTK: I PKA4F .;.- &SSZ.A.

~~ 11 .086L007

.018t.00=2~:-. 016!. 00 ...... 2~-

TYP R.005

' ! . co:

_,....+-1+·=.=.o 35

+ o .OCBRoF I .010:.oo

I •. ,. •oo· ~· .196 "

REF .005

.OSODIA

~ u.' WO rrN

DETAIL ·9-I :.x I

f 'i u

.. s ;;,. -I I

t ,, •• u ~ I ·• ~ "

'u C) .J

'" iii 7 (!j

lll lU OJ

"-,,, ([

... t1')

"' "' "' ··• I

·~ "' I ,_ 0) D

~National ~ &m'ltcoRcftictor

I High Performance Bus Product~

.-ti::i. ~

Siuclc Ended

Von---~

VoA----

DiKerentlaJ Sien"

Differential Waveform

VoA=VolS tVDiJT.

,. -. ,.. I VoA· VoB VoA·VoB ·u = -t0810V = +ttomV

VU ...._V.;..;t_h __ _,

--1--+tlllnlV

I 0 100.V

Vod:VoA-VoB

+400mV

OVDiK.

·4"n1V

• Small signal swing eases rapid switching speeds

• Differential signals provide noise immunity

TABLE 1. Estimated number of Gates required for a 60 MHz 3D-Flow chip assembly (32 cells)

30-Flow with 32 Functions/ Gates/3D-Flow Gates/ cells/chip.

Unit Logic Function Chip cell Unit Gates/Chip

30-Flow core Register File 32 3,072 11,422 98,304

ALU 20-bit 32 900 28,800

Program RAM '? ,,_ 6,400 G. equiv. 204,800

Prog. RAM control 32 250 8,000

Internal Host Interf. 32 800 25,600

Additional MAC 27-bit '? 4.500 4,500 1144,000 .)_

Interface to Look-up Table 4 6.144 G. equiv. 6,144 24,576 Calorimeter (256x 12-bit)

Host UART I 1,000 1,000

Interface

Output MUXOut 24 250x4 1,182 6,000

Direct Out 136 182xl 24,752

FIFOs Top 8xl2-bit 32 192 G. equiv (xl) 752 6,144

Control Top ,, ,,_ 80 2,560

NEWS 2x 12-bit 128 120x4 15,36)

Control ... ~n;x 24 200x4 4,800

30-Flow w/o 13,356

CAL interf.

w/oMAC

30-Flow w/ 19,500 450,696

CAL interf.

w/oMAC

30-Flow w/ 24,000 594,696

CAL interf.

,_. .:::. ""

l'rocc1Jing l~lcnk•Ul

Ccll a1ul chi('

1ll'!lcri11 tio11

111!1' w/o MAC w/o SCAN

FEI' .,.,,,

MACw/ SCAN

l'El'wl MACw/o SCAN

FHl'w/ MACw/ SCAN

l'HI' w/o MACw/o SCAN

Flil'w/o MACw/ SCAN

FEI' w/ MA<~ w/u Sl'AN

FEl'w/ MACw/

Cl'U c.1uck

............. ~ .. ,y frcc1u r11cy

in Milz

llllllt_.llU 120 uCMOS

....... Nmnbcr 110 ur gales ...

S11Ccol Ou1p estinu,.cd

in IM fnrc11\·h

Mii• (pl') Cl'll

240 10 X.X,JlJl.Jl 4RO IO

2411 IO •IRll IO

240 10 480 10

240 Ill 4RO 10

120 10 240 10

120 IU 2•IO Ill

llll Ill 2411 Ill

120 Ill 240 10

'l11hlc I:

'liilal Cosl- Cost per

Design l.ayuul l'abricali 'ICsl '(hlal

Pik.:ka ("usl- ( "usl- uu C"usl- C~ust-\VCCks chi11 ror

1u11uhcr Power (Ill d1i11s ~.IKHI unil ging Delivery Delivery Delivery Delivery

of 1!,HICS (wall) lcstctl a1uf pnuf. I >ic-(pins) SxlK- SxlK- SxlK- S•IK-

cslin1alctl tlclivc .. ctl) dic+p11l·k \Yccks \\'CC kl$ weeks \Yee ks

S•IK- ($)

X.JlX,JlX.X l.ll 447 x.xx.-aa XK.x-aa >txx.-aa xxx.-na xxx-aa xxx-a;aa y.y Cl'GA yyy-hh yyy-hh yyy-hh yyy-hh yyy-hl• yyy-l•hh

4•17 Cl'GA

441 Cl'GA

447 Cl'OA

447 Cl'liA

447 Cl't1A

447 I 'I'< lA

44'/ Cl'GA

~EH r·,.~,. ... ,.,u.., L:~vel 1 '"s~~" :(BO bo.:1tdsj :>ne Board· ot1 th'9 Pro1ramm

1-6 .i::i. "'1

SSCL 10/23/1992

- E\ . - el~t·trons - Jet

~ " ,. : E1 ~ ~ I ; 3D-now

~ ... I I B 1-~-1 I

~a I I I

i · I'·~·· I ~-

.... I I I l B I ~.

, ••. l Ttia•r with ~D-now tti,elllit1d Array. I

' .... < t ...

E1 ' 1 I t 1 011·l1r.11l I

' I~-"" I J l l="ll:l~r

I t ... . .

I ~t!l:~lver

I a I '

Cll"'tUltry

I . --· . ...

1 .... .....

' .... ········ ............ ·

Ro.w do.·to.

One o fiber ea.ch Tower ~

I . --

-I - . - . I . -0 0 0

_O I --I -~

pt1c11l For Tr1gg~r

·'' ~ .ll

1 Gb1t /sec/fiber

-Dario Crose'tto

- Al ~o r.:th ~S St'""'.,.laf,·o,. On 41 erosr•"'ll'l .. lp\~ proe4SS•t'•oltr•~

-· T:>e$.:~n o+ s~,te.M. .:1nJ c.~.:.-:>

R~C~tTec-rvRe

B~~tJ A rroio t:sr& Ctll r ,.,._ .,..,_..~-- ,..._. __

R &b eost: fu"' i 89 K. to .Z. 7' I(.~ 148

Presentation by:

W. Sippach

AM-TEST

L DllC 'l'/H - CHARGE SHAPER -24 llI'l'S . INJEC, AMP (ACU!.IN) _, I 'I' I I

'r'·~ 4 3 2 l

INPUTS

liCl.J( . HP82000 6

WADO '-- AM MEASURING 6

VECTOR RADO

GENERATOR 2 RD/RS'l' SYSTEM

.__ OUTPUTS

' 3 2 1

J. AH)/

' II HP-9000

'-- AOC llICRO•VAX WORK STATI~ 14 Ill'!'

~ 5 MHZ OP•LNX DATA

. -. (1211IT CHAN 30 MHZ)

-I c~v. OVM .___ .

24 llI'l'S OIG•SCOPE ' ~

-. DC CCllPAR

1.- ··SOURCE

TEK•11402

. ~ (' \ .... ,

PRELIMINARY RESULTS ON A NEVIS ANALOG ~ORY

GOAL: DEVELOP AN AM FOR HI PRECISION CALORIMETRY FOR GEM AT SSC.

a DIFFERENTIAL , 200 CELL CHANNJ::Ls. 60 MHZ SAMPLING , SIMULTANEOUS READ-WRITE FOR PIPE.LINING. 5 MHZ HUX READOUT TO ADC'S. 0.111 CALIBRATION , TWO 12BIT RANGES FOR 16 TO 17 SIT DYN. RANGE.

TEST'ClllPS: ( ORBIT l.2U J:)Ml)P N WELL PROCESS )

l 8 CELL 4 CHAWraL , INDEPENDENT R-W , NO DECODING - FINISHED

·2 48 CELL 4 CHANNEL·, INDEPENDENT R-W, FULL DECODING - FINISHl:!D

3 SAME AS 2 WITH OUTPUT MUX AND BUFFER OUTPUT TO ADC - DESIGNED

4 SAME AS 2 WITH PUSHPULL W-ADDRESS LOGIC TO STUDY NOISE.

PRELIMINARY RilSULTS FOR 4 CH 48 CELL ClilP: ( 12 FORSIGHT TINY CHIPS )

l DC LINEARITY ( MAX.DEV. ST LINE ) O.l %

2 DYNAMIC RANGE· > 3 VOLTS

3 TOTAL PEDWOTAL NOISB < l MV,

4 PATTERN NOISE DEPENDENCE: DC INPUT VOLTAGE INVARIE!lT WRITE FREQUENCY • ADDIU:SS PERMUTATION ?

-· 5 CHANNEL caoss-TALK ' - MOS'l'LY THRU PACK.

6 INDUCED NOISE FROM READ TO WRITE PROCESS

7 • • • WRITE • READ •

8 DROOP

9 AC LINEARITY

. 10 6 V, POWER VDDA 5 V. • VDDD AT 60 MHZ

< l MV

> l MV OF BINARY NOISE

< lMV I 10 MSEC.

FUNCTION OF DV/DT

15 MW ll MW

ll ALL ClilPS WERE FUNCTIONAL , AND NONE HAVE BEEN DAMAGED DURING T~T.

SOME CONCLUSIONS. : · ---,l,-..,B.TIUU<i EDGE: CHANGES ARE INDUCED THROUGH INPUT PAD TO SUBSTRATE

Cl-..PACITANCE ,AND NOISE IS Al>DED BETWEEN THE SUBSTRATE AND EXT. CNJ:>,

OUR DIFFERENTIAL READOUT REJECTS COMMON MODE NOISE , BUT HAS TROUBLE WITH VERY FAST EDGES,

2 AG LINEARITY IS LIMITED BY THE WRITE SW RES. DEPENDENCE ON SIGNAL VOLTAGE TIMES DV/DT.

3 WE SEE SYSTEMATIC EFFECTS FROM SEQUENCE PERTURBATIONS. THESE STUJ:>IES :REPRESENT OUR MAZN TEST EFJi'ORT,

. . . , . ,t' . ": ~ ·•· . ~'

··.-! ·, ~. •• • ·., ,,;< < ... ·, ,• ..

·.~,, .. ~;! ; ...... . ,·,:.,.

•·. "

··~ .. '•, ....

'' ·~· .,. I/,

" .; .. · ' '"

.. ··, ..

·•.: .

.,. .....

. ,'.-·1 .. -~ .•:,

. ~··

·, ' •' ... '

:.-: ~:: .. ,.

.. \ ....

' -153 ... ' . ... . .. ....... ~ .............. ..._ ___ ,

ll~U2 ~I~ITI~IN~ OSCIJ.l.OSCOiE l~;3.7,Q.i~:3.~l.~y:3,3 li&te;. 9-o.co:-92 time; 7;0:1.:3.9 ~~\i\llle~t IJJt ;po:i,o;.no

3V 'fj~~l~·3J;W1···"····c·:~!!~~;}!l~~~-~~-/i;1.~;l~I . . . ' . .. ,. ' t f ' I o f t I 0

I t I t _I I I I I t I I I I I t f I I I I I t t I I I I t t 0 I I I I f i o ' ' ' • I t I t I o t t I I I I t I t t t I I I t I I I o t O o I I I o o o I I o t I I I t O I I o o •

: )~,v. ·~ .S~~ o:i,.(.i. _,IC c~r~' ' -~ ' .-, • """ • I /I • ~./'"' · • · · •; • • • "• • • •; ·~'tl/'"°P. •r'i:); ·v-~;/' ·7r#r~ '/,"" .....

t I I t I 0 0 t IO t t t O t t t I I o t t 0 f I ' ' t ' t . . . t I I I I I I t It I I I It O I t ' t to I I I . .

l .. . I I t I o I

......... :v,~ .. ~ ... Vc~'I .... i. "ir.~;i~~/.. ~ ... : .... : ~: . : : · .. : • I I t I . ,,.: .: ''. ,·,;,: ••.• .. , .. -.:.:.-.... · .. : ......... : ., ....... : ... .

I ' • I I ' I o " I

• •

.. t I t· I ,I I_\ f I t I I t t f I t ~ I 0, t I I I t

'.~; •!:~ ;'.''..::;··:::·~· '· . . . : .. '

~:.;,::. _ ............. _"""'"-·------.. -----··-· .. ·-------; ........ • •

. . . . . I It t I It I I I I I t I It O I 0 O I 0 t I I 1-1 I Io I I I I I I 0 I I . . . . . . ... ,, ......................... ,, .. . . .

' .. . . . . . ff f •. t t t t_t t t t • • o t It I I: t t t O O ",'I I:.• It It It t O

0 I ' o

. ~ •· . ······································ . . . . . . If It 1111 llO,IJ Ill I l tt 'Io 1 o t t' 'I'• o' It tt o '.I If''

I t ' ' ' ' ,I I"

. ,·-· ·~ I' I ' . t ' t

t t.t'' t t t It t ft I I I It It It I I I I I,\ It It I If It IO I I I I

I , t ..... ' t·,.'.''. . I . . . ": • . '.

-2V ' .,., -17.~~~:.: . .~"-;',•;;,:; Sµsl'd~v .. '.. :.1]i1·

'·· • • ."t , . ' • If t ... 0 t 0 I O t"t I I O It t If. I I I f,I I I t IO t It rt ff,t "'I t't t t I It t t I It t t I tt_tf,tlf II It t I I I flt It If It It I I l~tl flt ,.t .. '· 0

• ',f

l•'.:,' , : •:,'.t t o t '· I I • t' "0

t ' t 'O' &.·' •

'.'.'•· ,1

1 ,, .••. rr_. , • '·

... · ~_;~·~ ' ., : ... ~ : ... :., .. ,.:;;;,·-,.;, ... : ......... : ........ '.;: .... . ~,·~·~!·'.:!.:····.······:······.. ·······:···. -', ·. ·: . ,:.: . : : .::· : • : C!' ~ ,_,;_.,.,,....- .OY.'..:. ,- • : t I I

1 ' ~r,ii"y,--,•, .~'I-:: I I 7"ftr."'tA('9'.f/,..,., o t t: o 'I I I Io I I' I I'' ······. ••t:•········:······ .. _.:·· ······:···· :· .. . . . ' ·: . " . . . , . . ..

. I. , ·,' :. " • . I I •. '. . •· ..• ' . ' • • ' • '·· ! • •:'1 I I '" I I t f I I t

··1 I 111111111111tttttltltftftttlttlttto•tttt111••••111ttt :• ' tfl.tlftt"\·lll•llltlfllltlll f.l',l!l,ll I t I .' t I

··_500m_v.:· .. ·· .. ~;.~ ..... :.··.: .. ·.:.· ·. : ·.·. : .. · .. :·· : · · . . . .;·: I lft:llllltttt:l,lllttt•tOtflt1t1000011t1tlltlllO I'd l' v lll!tltt .. tffllttlflfllftt lftt_ll,f-tt •t_tttl~'fll

1 ~ I• o • ' " t

' . '•,, o• • . : I l ·~-.·: ' I . I °.

. '• ..• :·· . l.;....,:.,_ .. J. __ ..;. __ ;.._.1-~,...·~·---·r-:-:-----~':'"--~'"'.-::--:~----.. ---.... -._· .. ·· ..

.. ;. , '• ... ,tr

• • • .. • .

••'..• .. , . • • • ,. ... . . . . . ·······························~····· : ' . . . :

• . . . . ...... : ................................ , .. . I ' o '

AvsCRl) .. i:- ........ :

·.Mo.in 1A~d "'•

··. < ;'J:.O' ... . ..... <A·'··· ;;;·9,, hi/'.. · .. -. :"'f:I . ·J1../Fb J '''""'"'"'

'• • ; ·~ ·t~:.,-·.·,·:_~ .•

11402 DIGITIZING OSCILLOSCOPE date: 2-NOV-92 time: 11:23:31

lexp:3.7,dig:3.91,dsy:3.31 Instrument IDt B0.10270

3. 462SV· · ···:·········: ....... · ·: · · ....... : ........ ·: ......... : .. · ..... · :··· · · · · · · :· · .. ·· · · ·: .. ··· · · · · · : : tVEVGgm5-z : : : . . . . . . . . . . . . . . . . .................................................... ' ................ \. . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . .. . . . . .. . ' ........................................................................... . . . . . . . . . . . . . .

. . . . . . . : :,,,, ~ I · ur : -0 · : .................. . c...'1-18. .. '· ...... ~7.P. ........... m-~Y ............... .

. . . /div

. . . . . ' . .. .. .. .. : ....... V~·-, ... ·i:, .. -.u. ;,,;..;,.;;;; . ;;,,.::;'-7.:r:. l:J.C . : ......... : . . ..

• ~ iV. • -ri ._,, ......... '-' • • I lfl'. · . . J . . . . . .

: lilt()#/ ~ /, ~'fr(?[ V~"i( 0° ~:. : . . . . . . . . . . ' ........................ _,_,,,,.,_, ..... _ .............................................. - ................................................................... ._ ... ,,_, ........................................... ······················ . . . . . . . . . . . . . . . . . . . . . . . ' . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ' .................................................................................................. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . .. ..... .. . . .. . ......... .. ...... ... .. . . .. . . . .. . . . . .. ... . . . . . . . . . . . . . . . . . . . . . . . . . .

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

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

i p.e~o ioC/IPW . . . .

:p..e~ftJu?fr :

11402 DIGITIZING OSCILLOSCOPE· 4&to; il•OC'l'-92 .time; 9:49;32

loxp:3.7,4ig:3.91,d~y:3.31 In~trw.mont IDt ~010270

fQ'"i~; twi.~l~ l:wL\!iWJ: i•,,G~1·111

11-:i 95 · •.l.Vtr.'<i~ .... ~:;:$. .... !.<!h'rf/,t. .I.~ .O:'lt'/l/.Z \~ .T.:4."~ :1./rl.£a. ~. rnv · .·· ·· .<.;.;.· • · ·~ • ·

.:; . . V'~s:v,'~/·V,~~ c/~~H~ ~ 1.rpz ~~rP4 . : ., ,.•.' • f • t I ' • ' •

' ..... ·· .. .·, :. ' '·

. :· ,,(i .• _,·· ',

~ ·.·-~ ·;. :~:>.·. ··.,:

/div ~

. ;,; ·-.._, I. .... , .

.. l/~~«'.:~~/~.: .. ~ .. Dc,: ......... '. ......... :.~ .. . Nlt~3;~;71~P.I· .. '··. :. . . .

. ,;·.: ~·::·~:\<~~~.': : . . .·:· ·; .·· ·. . : . ...... •'''.'' .. ' . . . . . •,. ' . . .

o • o 1 i I t o 'ft o ' o t 0·1 1 o o o I I O 0 • o O t ' I Ott 1 t < t t' I t ' ' ' ' ' ' , .. o o ' I t I I t 1 I I ' ' 1 t' I I I 1 ' I' ' ' ' ' ' ' ' ' ' '

' ' I • ' ,o f ' t ' . . . . ···. . .. ~·~·: tD · .: ...... ··.: : :~, f;: ~::.· · .. ·. : . : ' lo' .. • •,, ; . ·. '• ' I

t • • • • • I f I • ,o • o • .•· • t ·• • • t • o • • , • • • • •.I ,t 1· I, I·~ o I • • t • • • • • • · • • • • • • • • ' o • • • • • • • • I • ' • . ' . ·. . . . .. I

. . ' .

~~ 1'il1f/ 11·: ' .

111 111 I I I

! 1~11!!'i !i"I 1.i\11111i11 :1

. . . •,. · · · • J.iliJ.!Jw~,l~t I I •• .'. • • ~. I • • I' r i .. ·:: : ~ .. .,<:: ~ ~·. ~·- . . .

,. o,' , ' I t ':.1 "_,. ' '· . o'

w ll[ll''''.~ .... ?..0~"1¢.~.(~~·······~ --,..~,~:.~~/K-'""''f ... : I/. ; . ; ~. : : : j· .•• : ··: : .

. :.tr,i.~'d .".;"'~lrJ:f.~~:rv:.;~: : :: .. .;:'.· ·, ..• ~ o :-:-, o t •• ,'.-:1 •o • I P

' ........ ' )· \~J :-. ''\ .

" . ' ... •,

. . , ....

~~i ..... .,: .. jzp,n~·;;·Ji :r;.Ji ...... j;:-cw, r.1.~·z~·.k · IJ"' : . . y ; : ~· '; . . -1'\.'l. ' VI ;i_~

·, : ·.,. . .... : : : : . ' . ·; . . ... ., . ·; .. • . ' ' ' . .'.·,·,.·,.- •• ,'. I • '. ,' ' , • • ; , · .... ~· t ·' : I . . - . . . ' . , ' -.

o o f.f o, o t ft ff f I l·t I Io o 'o •·Io I 1 t I• IO 0 If I l, I; o o 0 O t O' I IO' O O I 0 O I 0' •'t I I I I',°" I 0 0' Io o' 1

.••. ' . . . . . ' . : : .. ' ' . ,, . ~ ;·. . . ~ .. ' . .'. . .. "' . '' ;·.·· :· ·., ·,•

f I • f.• I I • t I I . ' ' '

o I I I I I I t I ; I I, 1, I_ I I t I O : I I O o 0 I O I .• : O 0 ' !, O ! ~ I ... : I

' . : ., .·.·~: :~ . : . ·.. : ·' .! . '.' '· ' . ! . ' : r, .. . . . ,. . :

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

• ... '·.

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

' . . . . .. .

-·~· .. ;,. ......... ~ ·.·:...;;.,1, .... ·' ··~; ...... ~ • '¥' ·•

o~-) o~~;-r.----i-~--,.~--o C sample

0 __ _, wrsw

KTC sample +sub ~ KTC bua + wrsw partition noise + rstsw partition noise

rr 11e/f)()fl.,/~?

. t.J r1 If /'l-

5Y STtfn µa1se Bir Ir/Jc

p..tt)

._. 1:11 ~

1992-11-03 12:51:04.10

7 ....

6 .... '

3 ....

1 •-

R'1 S: b·'' JI -v

:__ '- ~

I I I I I I I

260 261 282 283 284 N= 46, H= 104

Pedestal Distribution (in MV) VCOM-NEVGEM3-X5-ADC48

(EOe:s-TA-C-5 ---------- -------~-- --

dc-nevqem3-adc. top

--1 H --1 (l) (.)

('\J 0 ('\J I

s:i.unoo :JOV :i. t.:J-1'\QJ

~ , ..... --Q)

(\') O'

'° ..... _, 0 > "-Q)

~ ~ lk ).

...... '1

-- ------· -- ·-·-" -··----· -__ ,_ - -

s-ng3-adc-16.topl 0 0 0

~ '<i'

~! 0 ~; 0 ' ...........

u L{')

~i i::::i C':I

I H ~ ::::>

::;a H 0 ~ 0 C':I

I H "' {"/) ..._ H \

0 (/J ~ 0 .....

~ H L{') ~ ~

a:S C\I ~

" H (,>

Pi C.)

.... H A

~ (J) L 0 < ~ i

""' """"

,.0 H If'\

(J) 0 u..

...-I 0

Pi L{') ";;:;:. -s t

a:S {"/)

'-0 <(_

~ 0 0 0 oO

C\l - - C\1- ""J\ I I ....

' ---.'.-

~ slunoo =>av palll.!1-.M.'eli \.....'

( ,' -' ( ,'

Sample after peak (S-AMU-ADC) 20~---r--Y--r-r---r--Y--r-r---r--Y--r-r-r--Y--r-r--r--Y--r-r-r-.....-ir-r-r--.--ir-r1

10 I- ::i:: I~ :c :::s

111:1 fl)

::i:: c..:> +> I ~ ::s :c P> . I- 0 p..

°' 0 0 I- I u 0 ::i:: ....

~· ())

:c r+ "d 0 II> :c 'O

+> .... +> .... """ -1ol

:c I :c ~ ::c :c

-20 I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I

1000 1500 2000 2500 3000 3500 4000 ADC counts

t:..1ve-/f(l.; ry tif 5/f/1/lU' / {6 r'- 5' A-r«.f', ;?E.f.4?,,--•

COMPARISON OF AH AND HI SPEE]) ADC SOLUTIONS

ADVANTAGES OF AH .SOLUTION:

·.1, REl)UCES ADC SPEED FROM 60 MHZ TO 5 OR 10 MHZ,

2, ALLOWS HUX OF ADC FOR 8 CHANNELS,

3, ALL SAMPLES CAN BE CHOSEN FROM THE SAHE GAIN SCALE,

.4, LOWEST COST SOLUTION,

5, LOW POWER,

6 ~ . ACCE:SSABLE TECHNOLOGY - CMOS ~

ADVANTAGES OF HI SPEED ADC SOLUTION:

1, NO ADDRESS SEQUENCE PERMUTATION NOISE AS WITH THE AH.

2 , NO CELL NUMBER RELATED CALIBERATION CONSTANTS AS WI'l'H THE AM •

. 3, POTENTIALLY BETTER t.INEARI'l'Y FOR SAMPLES WITH LARGE dv/dt.

4 , NO CELL NUMBER DEPJi:Nl)ENCE OF SAMPLING TIME AS WITH THE AM,

5, NO LIMIT IMPOSED ON LBVEt.1 LATENCY,

6, NO LIMIT I~OSEl> ON LBVEt.l RATE.

7 , NO t.IMIT IMPOSED ON TIME WINDOW •.

8. PROVIDES DIG;tTAL INFORMATION i'OR LEVELl,

COMPARISON OF AM AND HI SPEED ADC SOLUTIONS

DISADVANTAGES OF AM SOLUTION:

·1. SIMULTANEOUS READ AND WRITE REQUIRES ADDRESS SWAPPING.

2. REQUIRES A LARGE NUMBER OF CALIBRATION CONSTANTS:

CORRECTIONS FOR EACH OF 200 CELLS. CORRECTIONS FOR NON-LINEARITIES.

3. ADDRESS SEQUENCE PERMUTATIONS PRODUCE SYSTEMATIC NOISE THAT · CANNNOT BE REMOVED BY CALIBRATION.

4. LEVELl LATENCY IS LIMITED TO A FEW MICROSKCONDS.

5. LEVELl RATE IS LIMITED BY 'l'HE ADC TIME.

6. THE TIME WINDOW IS LIMITED TO ABOUT 5 SAMPLES.

7, INFORMATION FOR LEVELl·MUST BE PROCESSED SBPARATELY.

DISADVANTAGES OF HI SPEED ADC SOLUTION:

l. REQUIRES 60 MHZ 10 TO 12 BIT ADC.

2. REQUIRES DYNAMIC RANGE SPLITTING.

3 • NO SHARING OF INPUTS WITH 'l'HE ADC. .

4 • HIGHER COST.

5. HIGHER POWER.

6. DEPENDS ON LESS.ACCESSIBLE TECHNOLOGY (BIPOLAR OR BICMOS l•

C) \ ~ .....J

rt; ..........

eel ~ \'\\! \ '!-+ 0 .c ++

++ CCI ~ Q.() ++

;::l ++ I'll -t 0 ij. ...:.:: ~ ~ + + + C)

Ct\ .c + ..... + + ....

..p.J ++ + ...:.:: C)

~ c( Q) + + 0 0

+ + .q< -'-".....t +

~ N 'll

s ........ V'i ~ -

eel 00 ~ ~

\.!... ~J

Q) 0 'J ~

eel \J{ .c ~ 00 's\ ~

' j-. ~

0 ~ 0 0 0 0 \..J) 0 0 0 0 ~ 0 0 0 0 CZ .q< C"l N -

< ~ ~ "'\ ng3-s-adc20-pk-pl6.topl

"' J 0 i 0 ~ 0

"'- ...i-

H ,...-..,. H J u 0 H

~ < H 0 0 I 0

Cf) H C':> . _.. H

rn Ii.) ~ H i:i ~ '° H g ~ (l)

0... H "~ (.) V\

~ i::::i HH < (l)

~ ......, 0 \......_ 0 c+-t

0 '° C\2 ~ (l) ......... 0... s ~ '° Cf)

0 t-0 0 00

~ 0 0 0 - C\2-C\2 - I I

~ ~ ........

s1unoo ::>av pa11I.!l-.M."EIH ,, '

~ ng3-s-adc20-pk. top 1 ~ V\

H '<:!'

- ~ u H

~ < H 0

Cf.l H en ~

_...' ~ Ill

..!x: H _. i::

'° ;:I

Q) H 0 r)l (.)

~ u ......., HH 0 ~

'° H/:j < ~

H 0 CC'\ Q)

.--1 0 C\2

'° ~ Cf.l

0 0 0 0 oO ~

C\2 ..... ..... N-I I

' slunoo ::>av lU-M.'9.I ) ~

' -..J 166

1 ~ ng3-s-adc20-pk-p 16. top 1 ~

'° . N) 0 \I\ "- 0 0 J ~ "<;!'

H ~ ..--... H ..... u ~ 0 H

< H I 0 \J 0 r:n. H 0

~ ........... (."')

H ~ rn

1l ro H ...... Q) i::

~ ·H ;:I ~ 0

~ CJ Q). H (..) (""'\ S-t 0 0 ~~ <

~! ....... 0 Q) 0

,.0 0 C\l

~ ........ ~ s

?:: ro r:n.

~ 0 '1 0

0 0 0 0 oO

~ C\l .... .... C\l-I I ........

slunoo ::>av pall!i!l-.M.'BCJ

·. 167

fZ Cl ~ H

H l!l ""' ....

~ ~ ..:I

~) ~ -~ ~

\....;

'1 ~ ~~

8 Cl) <

< l:'oJ

SE v ~ H IQ CJ

t;i "° IQ CJ ii ~ N Cl·

.... < 0

"" :! IJ ~-

· . · · ADCl

'>--I '1'/H 1---------' '1'/H J . r-LE j,...---....._ ___ __.r

CONV.1 L K u., oa )II

10 Ill'~ ')---1

t1 VX . fJ-k# f/fl · 5 /)c..; ?T/ ,,JC..

ro p.eotJcu A.Ju. iJF Aile~ ,if?V! A.)(J iJ/7"5

LB J ADCl

ll'llUIA

1-· 1-c 50 llS

60 llHZ ADC

'l'HREE 16. 6 ns PHASES 'l'O mCREASE '1'/H AND RANGE DECISION 'l'IME

;1UL 11/l/£ f /f/6t!"S Tv j/(LfJJ1 /Je'

110/lff st:r vp 71,..., 8 r()fL rl-ftf ~.

'-.)' /Vle7A-- s7 tf'Dtl:f t-lf-7CH s Tt1-rc .s ~P A-t,,.J o h I)/\¥ J //L-¥-Y/ ,,v? Tl /'"t C:-

{10(' THtf . Vf;{

'- I '--

CURRENTSPLl'l'I'ER BASED Oii FERllll.AB SCHEME

.. " ,.(,

" ,.(, ..

lQ .... -... ~

i--·-1

.J.,.;'{Cff'-11710,./ . OF 5/(J~ff(... f)tlfll ;.J (.,,.

5,/Jn/ll#fr J;vTE/LtJ,4-t.. i/S l/o C/ri-tr .5Al"'ljJt,,1~fi-..J /IJ 5rn /lVr/ l/,fjl-J>tv'.i;J/l cF .

fJA5Cp %P..hlt.AIJ 1'71

COMPRESSION

AMPLIFIER

16.6-na DELAY LINE

ENCODE

LA'l'CH

60 Hilz

SEIJlC'l' AMPLIFIER

PIPELINED RANGE SELECTION

60 Miiz

SUMMARY

AMO DEVELOPMENT:

1. OUR FIRST GOAL IS TO FULLY UNDERSTAND THE SYSTEMATIC ERRORS RELATED TO ADDRESS SEQUENCE , ADDRESS SETUP TIME , AND COUPLING

. BETWEEN READ AND WRITE PROCESSES.

2. THIS REQUIRES BUILDING ADDITIONAL TEST CHIPS TO SEPERATE SUBSTRATE NOISE INDUCED BY CURRENTS THROUGH THE PACKAGE LEADS , AND NOISE INDUCED BY ON CHIP SWITCHING.·

3, WE ARE CURREN'l'LY DEVELOPING AN OUTPUT BUFFER AMP TO EASE THE PROBLEM OF IN'l'EaFACING TO THB EXTERNAL ADC.

4. WHEN THESE ISSUES ARE RESOLVED WE WILL BUILD THE FULL SIZE , 8 DIFFERENTIAL , 200 CELL PROTOTYPE CHIP THRO ORBIT FORESIGHT,

DIRECT DIGITIZATION AT 60 MHz:

1, WE ASSUME FOR NOW THAT INDUSTRY CAN SUPPLY 10 BIT 60 MHz ADC'S WITH AFFORDABLE COST AND POWER REQ.

2, WE WILL DEVELOP A RANGE SPLI'l"l'ER AS AN ALTERNATIVE TO THE AM , . BASED ON AN ACCilSSABLE ( TO US ) TECHNOLOGY,

3, WE ARE PURCHASING DESIGN TOOLS DOM TEKTRONIX FOR THEIR CUSTOM PROCESS.

4. THE ORBIT 2 BICMOS PROCESS ( USED AS A DEVELOPMENT TOOL BY FERMILAB FOR THEIR DIGITAL P.H. ) IS AVAILABLE TO US TJ;mU FORESIGHT. ·

TEST MODULE DEVELOPMENT:

WE WILL BUILD A MULTI-CHANNEL AMO - ADC TEST MODULE BY THE END OF '93 TO TEST THE SYSTEM FEATURES.THIS WILL INCLUDE CONTROL , DIGITAL BUFFERING , AND OUTPUT LINK.

10.0 I

Ill .... .... 7.5 Q)

u .... 0

I- 1-t '1 Q)

.:::. ,D 5.0 El :::1 z

Sample at Peak (-Pedestal)

2490 2495 2500 ADC counts

2505 2510

~ Cb P> ::i II l\) ..... co -..J

0 (JI c..:i

Presentation by:

R. L. Wixted

. ·-·· .• o-. --........ -. ---- ---- --. --- -.. --.- ----- -.-... - -.· --"EQ\)1 iie: M_e:-tJT s '_ 0Pnci N .s

, - ------ ---- -- -- -- --~

- - ---- - -------- - - -·- -

___ _ ---~~ ___ (2e: ~l>_o l>T "Se*\t:;MftT~l~- _ _ _ _ ___ _ _ _______ _

- - - ----- -- - -·----- -- --- ------ ------ --~------- -----

(,st f e_cp.,,.,"""-~ J-.:::. ".Su.•'\'\\\-\ <>.I/' 1

. __ -··--· .. __ ·_Avµ/_ ""'ii~• JYtuJ ro V1 Q°"k

--·----·---·- ····------- -·

-----. ----r- -·~----·-· ·-·· -·---·--~..;;:;-- -;· --·- --· ... -·· . . ~ rl

--· .

-----_~f~~~l---~--~~R't-J-~_j)(t'\_~) ~--~ii 'C-5 ~ . .:l21Xc;<l!'_o)ico~2)---~-- _ · : · · n l I

-- -----~- --- ----- ____________ . _____ .(}~_j ________ • ___ ------ - --- ---- - -· - - - . - . - - . ---- - -- -

• • .. •· .. -· • • t •.•.

. . -· :·· ~--·---- -~~l~~~-~~~~;~~~~--~~- ~-- ~- _; -· -.. ---~ -------- --- --- . -·- . -. ·--·· -·-·· ·- -- -- - . -- ···- . . . . ·-· z. . . . . . -~ --~---·--- -~--·~ ~~\?f-)~-~:~d-t3~1~ ) __ ~ i.~S"~J' - - - ------- ------ --- --- ------ --- ---------- ----- -

- • -. n Spo-h:...J. ~tJ£. ~-- Va.1~ {fl>:).~- '" c..

--- -· -·---· -·_h~:f:~l~s~i :>1·:~-{i~;s·(-~) "I. 133

SL~V\vl _· A~~l.~i ~ . ~¥-

Fu-~ st.J£- StJ ~ -z /(,C. (To J ~ 1S4- { /o b'~ts ) 179

PROBABILITY DENSITY FUNCTION

1· . \

1/ \\ :· b'o+· 1 OL.....-L./~~~~~~:':-:.......o.·~~-~"---;:;',

50 100 150 200 250 O ELECTRON/ION PAIRS

_ =---=~~-~~--=-=~-~-p:t~_:p<i,.~cJ __ Lic.-o~~~~~~-) - _ J2~ 1-1s~i:

---=--:=--:-=~-B~tk~~['"~~~~ --. -=-: -- :---~-=-: ~~c.~ci.-"'rl~~ ---------:------- --=-?~J: -l~~{i : Ji-· --------------- -- ----- - . - -- - - - ---- - _'(. . ...

_ _ ___ : __ __ . _-_ Pe...L..~ A:l> r_~ _ z.o _

_ _ __ :: --=-=----e~ ... JC>v1- ~<:4 k

···--. ---- ----·-·---- ------------. ·---·------- -- - -- --- -- - --

t_l .. tA._!\l.\t-.l~/ e.4 __ ------ ~

__ _ __ S"oo ks/.s

____ _IO_bJ-s

--:-----:-St.t~e~~.~~r±rr-~ 'j.~-

u -:-_mm ; m! "hl::i~:---:ltz, ~-:_I_ .,----,-------_:__' __:_________ -----f-- - --- --

~--, -t- -

---. ; -+---i----C-~-_L__;_ ' - .

- - - ___ _.____ ___ _ -------- --

-------

->---'-~-·-~ c _:_S-/iJ--+-~

-Et-=--· ·--'---~ ---~- ....-------------- . - -:;;_------:o.V~f

----. ~-~-

~ . I I

-- . -l _______ :_ ~~-:

... ·>--4-1

h !,,'t 4/~ -

-------- -"-:'7-"------- --J.:. u..

---- -·-·------_---_- j=_ -_--

-=~~-----!---·

-1----· 1

---------·~

' >-----··--

---- ---------,--______,.- ----------- --- -1--------·-- -

·--- -----------·-:----:-----~·- - ·-·-

·------·-----~--------------

-----·-----------

-() ... . -:1 Keo..Jo.,...."'1

·-·------------------ - ---------"I:..:..

- - ·------·--- .......... ·- .,_,.... __ ~ ··- -- - -- ~------- .. - ~

--------------- ---

-- ----··------- .. --------------------------- --

.... , ... ,., ,,..,,., ...

Presentation by:

J. Dorenbosch

..__,~-I •UI _ _,

SUB-SYSTEM DA 1,A RA TES

DATA RATE SUB-SYSTEM CHANNELS EVENT SIZE @lOOKHz

TRIGGER RATE !' 1' '

CALORIMETER 82K SOKB 5 GB/s :c ~l-

IPC 446K 80KB 8GB/s Iv~

SILICON 3230K 20KB 2GB/s i~

MUON 518K 40KB 4GB/s ~ '1

TOTAL 4276K 190KB 19 GB/s 2c;*"

t7.tj6 ~- r;-7 71, ;:: l~b /~ /~!vvi ,.,._ ~I j !) ' ) lo/(!./.'''""

'If jt. 7 ~~J~ I ·• Ii

r -G~c .:

,_. 00 "'1

DAQ OVERVIEW

00 oo] 00 • • • • ' '. '

Control Network

'. ' • . ~ '

b' r/7/,!J b 1£b' . -trigger

supervisor

256x 256

• x I Gbps

• modular 1- switch

t: "' • e 1-= ~

~,,, ~ ~ I ~-;:::-...

' ennt ..... . .... I data data • • ct ! data

' colltttar colledor collector

i 0 i I ! 255 I

• w1···t. w1···t. • . , LI I I "-'-I • • · ....... I ....... •• • 1"··~1 -· ..... ....,11 ..... ...... -,- I • I I I

I I I I

I I I I I ........ ........ ~·······" ••••••• LIA/CLK/tbrotde

DETECTOR

storage system 00 supervl•or supervisor

• • • , r r

.. ... ..... I data • • • data ; distributor distributor

I • 255

w1···t. .-t···t. 1..--1 ••• ~ .. ~, .. ~ ••• 1~;1 ..... .,..,u ..... ..... . .,

FRONT END GROUP

:~ '--- L:J OR • • •

jMuxl .. OR

\ redundant

EVENT DATA COLLECTOR

i--" FROM ()0 FRONT END GROUPS • ~ •· •

TRANSCl!IVl!R

D . FIFO ~ • TRANSMITIER/RECEIVEJI

--- • PACKETBOUNDARY - • EVENT FRAGMENT BOUNDARY

I• i '

I• I I

PROCESSOR

TO/FROM CONTROL NETWORK

l EVENT BUFFER

l ll'MBl

4 .... I lGbpo l'lBER

I .... I (fO DATA SWITCll) NXM

MODULAR SWITC:H DESIGN r·-------------i ' I . . I . I I :

0 • . .

' I 16 x 16 I

' • • ' I • switch • I ' • • ' I I ' - ' . I ' ' '

' I --- .. -----------·

-------------------------------i r·---- ~

L-J 0 1 .,lo 1 -. ' I • • • ' I

=n • ' I "'' I

• • • 16X 16 at ~R 16Xl6 .llliiiWih.·h 9 9 S\\'ift:h

or /"'- lo

16 x 16 :r ~: 16 x 16

switch switch • • •

isl ..11s

• I . ' • I

• ' • I •

' ' ~------------------------------·------~

--------------------------~ r·---------- I

L__.J I ... 1 1-:

' I ' I

• • •

I 16 x 1(1 s,,·i1rh

16 x 16 switch

• • •

I • • •

• • •

~ .. ~ .. I ::: i :116X16 F :

1 switch : switch I •

• I • 15 • 15 • i

2ss i "'I I "'I I ! .... iss ' I •---------------------------------------

TIME DIVISION MULTIPLEXED SWITCII (256 CHANNELS @ 1 Gbps) IMPLEMENTING 65,536

VIRTUAL CIIANNELS @ 4 Mbps

EVENT DATA DIS'fRIBUTOR

IGhpo flBER

(FROM DATA SWITCH) NXM

EVENT BUFFER (1'1118)

PROCESSOR OMA

TO'FROM CON11lOL NETWORK

TRANSCEIVER

• • •

TO PROCESSOR FARM

D -FIFO /::). - TRANSMITTER/RF..CEIVER

--- - PACKETBOUNDARY - - EVENT FRAGMENT BOUNDARY

"' ~ -':! ~

~1:' ~ ~l ~

"" Ir Cl

\ ......

~ u Q ~

• • •

l'l.l Q Q ~

.!;. Cl

Vl I' ... ~ GI

• • • • • • ti.) • • • Q Q • • • ~

0 0 • 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0

= .-1 M z

fl.l Q Q ~

• • •

• •

tl.l Q Q r;i;;i

• • •

• • 0

0 0 0 0

rl.l 0 0 0 0

Q Q ••• ~

rl.l u Q 0 0 0

~ 0 0 0 0 0 0 0 0

• 0 • 0 0 0 0 0 0 0 0 0 0 0 0 0

• • • • 0 0

C'l.l 0 0 • Q Q • • • ~

C'l.l u Q 0 0 0

~ 0 0 • 0 0 0 0 0

0 0 0 0 0 0 0 0

Minimum Execution Time Calorimeter Algorithm (500 MIPS)

o.oo ~ <~ . . ;:t; . . . I . . . . I . . . . I . . . . I 0.00 0.05 0.10 0.15 0.20 0.25 0.30 - Total

TRIGGER SUPERVISOR

processor requests event (processor id)

send Lt id to processor

(+trigger type?)

remove Lt id from list

Lt Ace trigger pulse arrives

Inc Lt counter andaddLl id

to list

throttle the Lt trigger

DATAFLOW OUTLINE

N Q .....

~------------------------i

• I • I LIA arrives from

I+-trigger supervisor

data read from front-end into

event data collector

•• event data collector

builds event fragment and stores Into output buffer

• I • ~------------------------·

EVENT DATA COLLECTOR

-----------------------------------------------------i process

requests event

trigger supervisor supplies the event id to the processor

processor requests data from the appropriate

event data collectors

event data collector transmits the data

to the processor via the data switch

processor sends message to the

remaining event data collectors to discard data for

that event

L2/L3 processing

request remaining data from event data collectors

processor begins to write data to tape via the

control switch

-----------------------------------------------------L2PROCESS

Presentation by:

K. Freeman

. ·- . . . -· ~ -, ...... , . ' - .... - . ~ . -

o'f-2-.-a.~ ~ on 0 -f Tr-an ~ ~'\'N\J2.rt

.AC.NE/IC F1£LT-> Tt:ST

.S E"TUP + .McA-SU~E"Mt"NTS-.

·.J~ ~~ ;J.~C>~ {c./£,J~L~ ::6>-- ,,._. 3 k ~ {FFs)-~~ ~-

. ~~~~~ ~~~: - P~ ~!;)_OJ 60; 6C>~ RMS

-~,,~/~~ - B~)(TE"TeNAl J ~ > 3 /<. ~

-.J~~~~.

C..CJNCLIAS/(')NS (£ARLY REgULTS)

·a:.l~~~;~ ~ ~ ~ BFkreR.NA-L ~

~~ ~)·P~~~ =?>~~~~ j~~-~~~.

~)·1~~~~~ faitTvtAL tNDfA.C.TANC.t ! 1 1

• B~ ,._, Jf Do~'~~;;{:$- Be"' =:!;>) C/£ .,JL._pl l..u·~e.. o ~.

-~~- ~: -P~~~ -~~,~J_ -~~-.~~~~~~

-J~~o~l9f-~~-

=== ...... ...... zzz ...... 5:88 _,. ;i:: ---""" """

FUTURE IA/DR/< - R.£FINEA4t~lvT -1-cXFAl"StoN

L~ -~....i.:ir"-~ ... L~~-~~-.....,._._~.......-e.

-~~~-~~~ -~~)~~

• a,,lL ~ RAS'~~_..,.,~~~

·~~-~'~ ~

• ~ ~ ~ ~4HJIA~-'-""l

~ . ~~""~~~~ ~~.

"'"'"' .. __ WWW !:: "'"'"' 000 "'00 _.., _..,,. ...... ---c:." r. .........

I . • ' , ·~---> _,

"UPRIGHT"

POSI TIOfJ

Fl c. U f.~ l cv

"S IDEWA'r'S"

Fos JTt(;)/V -r\ :._ l _j -~- ·"ct. . 4 ! ' • i: :·- __ ,_ '--

' . . i I i• . Y'

j EXT~N4L

L/N£S

E:X TliRNAL._ FfE'L D _ FL I.I)(_ LI f./ t:s_

~t_R, C':f_ f,,., AT! N ~ ~~)(_ tNf?L{CE"P /3'r PRIM4R_t'

CU.R. Re NT ! i

I ~"-LATIN <i i' FLIAX REINFo_f?CeD SY I COl.lPLEO FL I.IX.!

!=LUk CAf\/Ct=l..L.E'D er co1.1PLE0 FLl.fx

FIELD FLU.><

FLUX. E'S REIN/:oRC~

FLUkeS CAtJCEL

...... ......... ..... ~ ...... :::: ...... ooo .. oo ......

CU.RR.ENT/ND C< Ct"D FL I.AX LINES

Ff <i. uR r J.. c...

/.~r~---. • 1 ",, • VT -._

e I cR N A-i, Ft ELD! FL W>c Ll111e.S j

I i l ~

"f(AT" ORIENTATION

~.---_:;: 0 a'----J

~ _,, - '~ 7 ~ ·-cs ?)

~IJJ\L

...... ':'':'':' ;;; .... -,,_ oo .. 000 ...... :z:::c::c mmm mmm ........... ......

-... ••• ---N,. ~~~

I _·r;71: -1- -1 -~- ~: -

.... ' <-! ~"¥ i () } ~ -<I" n..

r~~ .. 0 S.-i:i

"'- ol "' I

() ~ 0 !

.... Q.. IJ I

"' u "'t ~ '!: :::.

':>.! -tT' ')

'~ _) ,.

Sampling Resistor Voltage, (Vpp)

Increase in Exciting Current due to External Magnetic Field

(120 volts, no load)

0 1000 2000 3000

External Flux Density B (gauss)

p;c,. lJR£ 4-

Upright

Sideways

Sampling Resistor Voltage, (Vpp)

Increase in Exciting Current Due to External Magnetic Field

(60 volts, no load)

0 1000 2000 3000

Fl~ UR~ S"

El Upright

Sideways

Sampling Resistor Voltage (Vpp)

Increase in Exciting Current Due to External Magnetic Field

(30 volts, no load)

o.+-~~~::.----.~-.--.-~...--__J 0 1000 2000 3000 4000

p/G, uR· E 6

GI Upright

Sideways

Sampling Resistor 30

Voltage (Vpp)

(upright position, no load)

0 1000 2000 3000

f/G, UR£ 7

Upright 120

Upright 60

Upright 30

(sideways position, no load)

oa. ..... ===!!~~~~~~~~~~~~........J 0 1000 2000 3000 4000

External Magnetic Field B (gauss)

Sideways 120

Sideways 60

Sideways 30

(flat position, no load)

0 1000 2000 3000

Flat 120

Flat 60

Flat 30

I 30-Flow Processor for a Calorimeter

Programmable Level-1 Trigger I

I D. Crosetto

October 1992

3D-Flow Processor for a Calorimeter Programmable Level-1 Trigger•

Dario Crosetto

Superconducting Super Collider Laboratoryt 2550 Beckleymeade Ave.

Dallas, TX 75237

October 1992

SSCL-Preprint-164

*Presen1ed at the 1992 Nuclear Science Symposium (NSS), Medical Imaging Conference (MIC), Ocl. 25-31, 1992, Orlando, Florida.

toperaled by the Universities Research Association, Inc., for the U.S. Departmenl of Energy under Contrac1 No. DE-AC35-89ER40486.

30::..Flow Processor for a Calorimeter Programmable Level-1 Trigger

D. Crosetto Superconducting Super Collider Laboratory*

2550 Beckleymeade Ave. Dallas, TX 75237

Abstract

The types of detectors and the physics involved in present experiments are reaching a level of cost and complexity so great that it is preferable to implement a programmable trigger solution at all levels rather than a system realized with cabled logic. Experience demonstrates that fine tuning on the trigger is often achieved only after running an experiment and analyzing the first data acquired. A Level-I trigger is required to identify objects (particles such as electron, jets, etc.) with programmable algorithms at 60 million frames-per-second. These requirements have led to the design of a special "3D-Flow" processor that, together with a special pipelined parallel-processing architecture, allows a sustained data rate of 60 million framesper-second. The 3D-Flow is a data-flow processor that can be used in one-, two-. or three-dimension array for high-speed signal-processing applications such as identifying objects in a matrix in a programmable form. Feasibility studies demonstrate that with present technology a 3D-Flow chip consuming 8 W, and accommodating 32 processors at 60 MHz can be built today, and that the same chip can be built one year from now at a 120-MHz clock rate.

I. lr-."TRODUCTION

The Superconducting Super Collider (SSC) is being built to study high-energy physics. Every 16 ns, proton beams will collide and the panicles produced by the collision must be identified and studied. Many detectors will be used to detect and identify the panicles. The calorimeter is one of the sub-detectors used at the SSC. Two proton beams will collide in the center of the calorimeter. sending particles to the calorimeter towers in the barrel and end caps. The amount of energy released in the collision is detected and then transferred through channels to digital processors, where the identification of particles is begun in Level-I triggering. The triggering mechanism must be able to rapidly reduce the amount of data by discarding unimponant data. The calorimeter will provide Level- I trigger information regarding electrons, photons, jets, and missing Et (such as neu-

"Operated by the Universities Research Association, Inc., for the U.S. Department of Energy under Contract No. DE-AC35-89ER40486.

trinos). 1 The starting point is the off-line algorithms that require milliseconds (ms) for execution. The challenge is to find a given "system architecture," "processor architecture," and its "instruction set" that provide the best and most suitable (to the component) conversion of the off-line trigger algorithms to a fast and simple "real-time" algorithm that will still have high particle-identifying efficiency.2,3

II. PHYSIC REQUIREMENTS

With a programmable solution, it is possible to use the same electronic (commonality) chain for several experiments. For this reason, and because all physicists do not accept a specific type of trigger algorithm, a programmable solution is highly desirable. As an example of programmability, many trigger algorithms have been implemented. Among these are finding a local maximum, calculating cluster and transverse energy, comparing cluster and partial sums to different threshold, determining whether an electromagnetic cluster is isolated from nearby energy deposition, and determining whether a 4 x 4 matrix is a possible jet. There are many conditions to test when making the Level-I decision. As an example of a programmable system, a few methods that will verify these conditions will be implemented using the 3D-Flow parallel-processing system array. Figure 1 shows how to check some of these conditions.

In order to distinguish electrons and photons, the "em" trigger tower energy must be greater than a threshold, the "had" to "em" ratio must be very small, and if isolation is to be achieved in Level l, the surrounding towers must contain only small amounts of energy. For jet identification, the sum of a tower matrix must be tested against a threshold. To distinguish neutrinos, the E, sum must be compared with a threshold (see Figure 1). An isolated electron should be recognized by a large amount of energy deposited in a small area (about 1 tower wide) while a jet's energy should be spread out covering a large matrix of calorimeter towers. Figure 2 show the flow-chart of the algorithm that executes two "em" sum + front-to-back+ isolation + jet finding.

Local maximum

Jet ~ finding EEEB

C > Ii for i = 1, ... , 8. • Threshold < L: Ii + C .

•••

Threabol4 <~&., + L:z., tm't 1•1

I ~ I 1, ~' "' ... I " Ill

• 1 I ~ t:;1 ~ !

s/ : ~ !

ii ' ~\~

FIGURE I. Trigger algorithms examples.

Ill. 30-FLOW PROCESSOR ARCHITECTURE

The architecture of the 3D-Flow processor is shown in Figure 3. The 3D-Flow operates on a data-driven principle. Program execution is controlled by the presence of the data at the five pons (North, East, West, South, and Top) according to the instructions being executed. A clock running at a frequency of 60 (or 120) MHz synchronizes the operation of the cells.4

IV. 30-FLOW PIPELINED PARALLEL-PROCESSING

ARCHITECTURE

Figure 4 shows the 3D-Flow processor in a pipelined parallel-processing architecture.s The program execution at stage I must not only route the new incoming data from the calorimeter to the next stage in the pipeline staging (stage 2), but must also execute its trigger algorithm. A graphical representation of the input and output results and the data flow in the pipelined architecture are shown in the timing diagram of Figure 5.

Phase 1:

Phase 2:

Phase 3:

set code

set code no

set code set code

output code

1 two "em" sum !north 1 x 2) > threshold 2 two •em• sum east 2 x 1) > threshold 4 "had" "em" l"°rth 1 x 2) < threshold 8 "had" "em" east 2 x 1) <threshold

16 Isolation achieved 32 possible jet found

FIGURE 2. Algorithms flow-chart. For example, a 3D-Flow may return a code 37 (1+4+32) stating that a possible electron was found, but it was not isolated from the surrounding energy, and that cell may be part of a 4 x 4 jet.

FIGURE 3. 3D-Flow cell architecture.

Stage ll /' /' Stage 3 )'

Stage 2 Stage

V. 30-FLOW CHIP ASSEMBLY AND BOARD

ASSEMBLY

Preferably the assembly of the chip should be in a good thermoconducting package with good power and ground distribution, such as the 447 Ceramic Pin Grid Array with 96 pins of Vee and Gnd. Feasibility studies demonstrate that with present technology a 30-Flow chip consuming 8 W, and accommodating 32 processors at 60 MHz can be built today, and that the same chip can be built one year from now at a 120-MHz clock rate. Figure 6 shows an assembly of 8 x 30-Flow chips on a single board, thus making it possible to analyze calorimeter channels relative to a 0.4 x 0.4.

- E, - elect,..ons

~ - Jet ~do.to.

OM optlco.I FIMr For ••ch T r ..... •

FIGURE 4. General scheme of the pipelined parallel-processing architecture using the 30-Flow.

l~I II Raw data Sta2!!: 4 Sta e 3 Staa!!:2 staae 1

;~. ,r----T----· ----- ----- ' 0

I had -· ·- 16 I Algorithm - had

ncy execUtlon time - -· I

\ """ I I 32 I had

I ". I --I i I 48

'< had , . .

Late ns

I ·- 64 < I • • . had

. I --I '7 BO ns

' • had I ·-

I • 96

' .... ns

I -- 11 2ns

' I • """ • --• • 1 28ns

,. .... ,. ·-•n 1 ....

'" rn had

" 1 eons

-- 17 6ns ,, had

1 19 2ns

Time FIGURE 5. Timing diagram of four stages of pipelined programmable Level-I trigger.

FIGURE 6. Board of the programmable Level-I trigger with 30-Flow array.

REFERENCES

N.Ellis,"EagleTrigger/DAQ/FEGroup,"CERNDAQ-TR-109 20/2192

D. Crosetto, "A fast cluster finding system for future HEP experiments", Nuclear Instruments and Methods in Physics Research, A311, (1992), 49-56.

D. Crosetto and L. Love, "Fully Pipelined and Programmable Level I Trigger," SSCL-576, 1uly 1992.

D. Crosetto, "30-Flow Processor Technical Specifications" SSCL-594, September 1992.

D. Crosetto, "3D-Flow Processor for a Programmable Level-I Trigger (Feasibility Study)" SSCL-601, October I 992.

Electronics Meeting - Brookhaven National...

Documents