SUPERCONDUCTOR AD-A248 .159' TECHNOLOGIES Ill~l 11I IIlTHE PROCESSING OF HIGHTEMPERATURE CERAMIC

SUPERCONDUCTING DEVICES

Contract No.: N00014-88-C-0713

Principal Investigator: Dr. James H. Long, Jr. D T ICARPA Order No.: 06268 ELECTE r'

Contract Dates: 9-1-88 to 12-31-91 APR01 19.

.S DFINAL REPORT

Volume 2January 31, 1992

Submitted by Program Manager

James H. Long, Jr.This document has'been approvedfor public release and sale, itsdistribution is unimitod.

SUPERCONDUCTOR TECHNOLOGIES INC.

460 WARD DRIVE, SUITE FSANTA BARBARA, CA 93111-2310

Phone (805) 683-7646Fax (805) 683-8527

92 3 25 067 92-07621

THE VIEWS AND CONCLUSIONS CONTAINED IN THIS DOCUMENT ARETHOSEOFTHE AUTHORS AND SHOULD NOTBE INTERPRETED AS NECESSARILY REPRESENTING THE OFFICIAL POLICIES. EITHER EXPRESSED OR IMPLIED, OFTHE DEFENSE ADVANCED RESEARCH PROJECTS AGENCY OR THE US. GOVERNMENT.

460 WARD DRIVE, SUITE F, SANTA BARBARA, CALIFORNIA 93111-2310TELEPHONE (805) 683-7646 FAX (805) 683-8527

BET MOTION SYSTEMS COMPANY

STATUS REPORT FOR

LINEAR ACTUATOR PROTOTYPE

10 July 1989

Contract No.: N00014-88-C-0713-S-BEI-1

contractor: BEI Motion Systems Company

2111 Palomar Airport Road Suite 250

Carlsbad, CA 92009

Accesion F. .

Principle Investigator: Mr. Jack Kimble rTIS CoI&I V&DTIC IAb

(619) 744-5671Ju~I~C nt Y....... ...... _..

By ......... . . ..

REPORTIING PERIOD ENDING JUNE 1989 A,,ib,/ ." '. : .'m .i "

Statement A per teleconDr. Wallace SmithONR/Code 1132Arlington, VA 22217-5000

NWW 3/31/92

1

I I. PROGRAM SUMMARY

I BEI Motion Systems Company has contracted to work with

STI to design and specify planar coil suitable for

3 linear actuator. If films are made with sufficient

current densities, BEI will design a linear actuator

3 or brushless DC motor system to incorporate planar

coil.

U II. PROGRAM STATUS

IBEI has completed the preliminary evaluation on the

planar coil and conductor requirements for the brush-

Iless DC motor. Results indicate that current density

is adequate, however, there is not enough cross section

of the conductor using the thin film approach. Pre-

liminary design was based on two planar coils attached

3 on each side of a central chill plate filled with

liquid nitrogen. Evaduation was further performed on

3 the ouput anticipated in thick film substrates 100

times thicker than the 1 micron thin film. Estimates

with thick film substrates yielded .168 watts output

which is still inefficient for DC brushless motor

design. Thick film designs appear to be the correct

Iapproach to achieve a high conductor cross sectionenabling sufficient power levels to drive the actuator

3or motor. BEI evaluation has included detailed review

and analyses of the following key areas: STI achieve-

ment of 105 amps per square centimeter of output in

thin films; planar coil actuator design requirements

5 including number of conductors and substrate size; and

anticipated power output.

III

II

III. ACCOMPLISHMENTSIProgress was made with the contracted work statement.

I However, there were no firm accomplishments as thick

films are not available for design consideration.IIV. PROBLEM AREASI

None experienced. A heightened level of attention

needs to be given on thick film approaches to ensure

continued progress.

I V. CORRECTIVE ACTION

3 None.

3 VI. GOALS FOR NEXT REPORTING PERIOD

I Continue low level research activity in accordance with

approved funding levels for the planar coil design.

I VII. FUNDING STATUS

I a) Expended: $11,547.56

IIIIII

I

THE PROCESSING OF HIGH

I TEMPERATURE CERAMICSUPERCONDUCTING DEVICES

Contract No.: N00014-88-C-0713

Principal Investigator: Dr. James H. Long, Jr.

ARPA Order No.: 06268

Contract Dates: 9-1-88 to 12-31-91

IFINAL REPORT

Volume 2January 31, 1992I

ISubmitted by Program Manager

James H. Long, Jr.

ISUPERCONDUCTOR TECHNOLOGIES INC.

460 WARD DRIVE, SUITE FSANTA BARBARA, CA 93111-2310

Phone (805) 683-7646Fax (805) 683-8527

I

! i ThE VIEWS AND CONCLUSIONS CONTAINED IN TH[S DOCUMENT ARE THOSE OF THE AUTHORS AND SHOULD NOTBE INTERPRETED AS NECESSARILY REPRESENTING THE OFFICIAL POLICIES, EITHER EXPRESSED OR IPLIED, OFTHE DEFENSE ADVANCED RESEARCH PROJECTS AGENCY OR THE U.S. GOVERNMENT.

I

. x-

oon

0.. 00~

0Lo

L.0 0....... Cc

CD U)

4) >

ii

R & D STATUS REPORTIDARPA ORDER NO.: PROGRAM CODE NO.:

3 CONTRACTOR : Avantek Inc.

g CONTRACT NO. : N00014-88-C-0713-S-AVT-1

EFFECTIVE DATE OF CONTRACT: September 1, 1990iEXPIRATION DATE OF CONTRACT: August 31, 1991IPRINCIPAL INVESTIGATOR: Amarpal Khanna

TELEPHONE NO.: 408 943.4226

i SHORT TITLE OF WORK: Superconductor Oscillator and PackageInvestigation and Development

REPORTING PERIOD: JULY 1991 TO SEPTEMBER 91

1 o DESCRIPTION OF PROGRESS :

i Attached Herewith

IiIIII

II

SUPERCONDUCTING OSCILLATOR (SCO) AND PACKAGEINVESTIGATION AND DEVELOPMENT

i R&D STATUS REPORT FOR THE PERIOD OF JULY 1991 TO SEPTEMBER 1991iDESCRIPTION OF PROGRESS:

5 Package Development:

Ten each of aluminum cases model SCP-1000 and stainless steel casesmodel SCP-2000 were assembled with RF feedthrus and 50 ohm thrulineon alumina between the input and output ports. The thin filmcircuits were fabricated in Avantek. The internal dimensions of thealuminum package will accommodate a 1 cm square substrate and theSST package would accommodate a 1/4 by 1 in long substrate. Thesubstrates are held in place with BeCu clamps and 0-80 SST screws.The clamping arrangement is used to allow for mismatches incoefficient of expansion between the substrate and the metalpackages. Standard SMA connectors are used with RF feedthrussoldered into place. RF connections to the substrate areaccomplished with a gold ribbon wedge bonded to a contact pin. Thecontact pin engages the centerpin of the RF feedthru centerconductor and exerts a constant loading to the center pin. Thisdesign technique precludes the need for any soldered connectionsinside the package. The lid is laser welded into place.

Experimental Method:

The ten cases of each type were welded and leak tested. Electricalmeasurements of SII and S21 were then carried out using the networkanalyzer. The packages are mounted on a 1/4 inch aluminum or SSTbase plate. The packages are placed in an insulated box and liquidnitrogen pouired into the box. Steady state temperature at 77 deg Kis reached when gaseous nitrogen is no longer observed bubblingfrom the surface of the packages. It typically takes about 20minutes to reach this condition. Temperature cycling of these casesfrom room to 77 deg K was carried out. A total of 75 cycles werecompleted in steps of 5, 5, 10, 30 and 25. The packages were testedfor leakage as well as Rf (S11 and S21) performance at the end ofeach step. The results of the testing are summarized below. Detaildata of the RF testing and leak testing at the end of 5, 10, 20, 50and 75 cycles is attached as appendix 1. Copies of thedocumentation on the two cases SCP-1000 and SCP-2000 is enclosed as

i appendix 2.

II

II

Number of units passed testStainless Steel Aluminum

# of Cycles RF Leak RF Leak

3 Start 10 10 10 10

5 10 9 10 10

1 10 10 9 9 10

£ 20 10 8 9 10

50 10 8 9 10

5 75 10 8 8 10

I Out of 10 SST packages S/N 10 and S/N 7 failed fine leak testafter 5 & 10 temperature tests respectively. All other partspassed the leak test up to 75 cycles. All SST packages passed theRF performance test up to 75 cycles. Final examination indicatesthe possibility of leak test failure due to poor soldering at RF

feedthru.

Out of 10 aluminum packages S/N 002 and S/N 003 failed the RFperformance test after 10 and 75 cycles respectively. All thepackages passed the leak test up to 75 cycles. The units were cutopen after the completion of the tests and the gold ribbons werefound broken. The failure appears to be due to insufficient stress

I relief.

IIIIiIi

I

Superconductor Oscillator Desiqn:

I Investigations continued, on developing low phase noise

oscillator using superconductor resonator. Main effort during thisphase of development was on developing a series feedback integratedoscillator. Some experiments were conducted to conclude theparallel feedback investigation.

5 Parallel Feedback Superconductor Oscillator Design:

A bipolar amplifier was redesigned at 10 GHz using commerciallyavailable Si-Bipolar device NEC647. This device can provide atransmission gain of better than 6dB under matched conditions.The second prototype built provided >4dB of gain and was usedas front end amplifier to the GaAs FET multistage amplifier. Thephase noise using this amplifier was still about -lOOdBc/Hz at10KHz from the carrier. Phase noise of better than -100dBccould not be achieved using the parallel feedback approach, itwas decided to continue efforts on the series feedbackintegrated superconductor oscillator.

3 Series Feedback Integrated Superconductor oscillator:

The series feedback approach (figure 1) is a more commonly usedconfiguration for designing oscillators and can be easilyintegrated with the superconductor resonator in the same compactpackage. The DC power required to generate 5 dBm at 10GHz isexpected to be less than 50mw (3V,15ma). The resonator is a highQ notch filter at the desired oscillation frequency. Theoscillation condition (S1I' X Ti = 1) is satisfied bychoosing proper coupling of the resonator and the distancebetween the resonator and the transistor.Ic

I si_

Su

III

5 Fig.1 Series Feedback Superconductor Oscillator

I

CD0

C -0 0I0

Jo.

700

(D 0-

-f C 0Cl 0

0 - 0

+ 4-

-H

.VVI I

+ 01

+1+

ILLWT N Q

030CD 0

Cua)

+

130aaCC

L) 0LO 0NI0

I Cl+I-)

____ __ __ _ ______ _ __ _ _ ___ ___ C+

En- _ _

cc cc

CIU

IN I- II * V

U I I I ____ ____ ___ '7ml ___ __ __ __ __ _--m T i , 15 I I

I i I I r- (-

~iI '

.1

1~-- 0

Li- I

I VUI

I C)

Tn _______ ___ ___ ___ ___ ___ ___ U1 INI I

C)'I I II \* 4-4

- Fr-

ii 4~)1 I C)I I *I[Q

1 I N )L _____ _____ I- _____ _____ _____

____ ____ I ______ __ IL\ N* I N~ __ __ 1 __ __ -F

- 1 IIII I

I 00 rlLa wH

* H ~0JJ* F-I <V-41 urn1 0

I

I

I Generation of negative resistance:

In order to satisfy the oscillation condition (S1I' X Ti > 1) thereflection coefficient Sll'(figure 1) needs to be greater than1/T1. Distributed capacitive feedback at the device source is usedto generate the negative resistance(Sll' >1) in the device. Thedevice used is Avantek M125 GaAs FET. A linear CAD program was usedto design the active circuit. Figure 2 shows the active device

negative resistance at room and cold temperature. The S11'increased by about 2dB from room to cold. The value of Sll' at coldshould be one to two db greater than the T! of the resonator.Two negative resistance substrates are enclosed as part of thefinal report. These substrates can be used to build superconductoroscillators between 9 and 11 GHz.

High Q Superconductor Reflection Filter:

The superconductor reflection (notch) filter was designed andfabricated at STI. The resonator was tested for reflectioncoefficient and Q factor at 77 deg K. Typical resonance response isshown in figure 3. The T1= -3dB and unloaded Quality factor of 3500was measured at 9.7 GHz.

I Series Feedback Superconductor Oscillator:

The superconductor resonator and the active negative resistancecircuit were connected through a predetermined length of coaxialcable.The super conductor resonator was packaged in Lne standardSTI package with SMA connectors.Active negative R circuit wasassembled in an Avantek standard DSO case with 50 ohm linesconnected to extend the circuit to the input and output connectors.The length at the input was adjusted to provide the necessary phaseshift between the device and the resonator to satisfy theoscillation condition. Fig. 4 shows the phase noise of the seriesfeedback oscillator at 9.7 GHz. Performance of this oscillator isshown in table II: Table II

Frequency: 9.7 GHzPower: 5 dBm

Pulling:.2 MHz @ 12 dBrPushing:.] MHz/V

Phase Noise: -95 dBc/Hz @ 10KHzBias: 5V, 20 mA

The oscillator discussed above provided important data that areflection feedback oscillator can be successfully built.Reproducing the above design is difficult because the resonator andfilter are in separate packages. The only packaged high Q notchfilter available, stopped normal functioning and it was decided tocontinue work on an integrated oscillator by using the same packageto accommodate active circuit as well as the superconductorresonator.

I

I4-1

(a 0

H fa

(~0 4J

_a 0

CD

C--4

I D

-4-)

I4C

CN. 04-i~Cf) 1-

rn- C

-r I

((II

I

I Integrated Oscillator:

Superconductor resonators were received from STI on substrates.The active circuit and a reflection filter were assembled in thesame stainless steel case. The length of the 50 ohm line betweenthe device and the resonator was adjusted to meet the phase part ofthe oscillation condition. The oscillator functioned at 9.47 GHzand had the following performance characteristics:

Frequency : 9.47 GHzPower out: 5 dBm

Phase Noise: -82 dBc/Hz at 20KHzPulling into 12dBr: 5 MHz

Pushing : 1 MHz/VBias: 4V, 25 ma.

One integrated oscillator at 9.47 GHz is enclosed as part of thisfinal progress report.Figure 5 and 6 represent the phase noise ofthe oscillator.The phase noise achieved was about 15 dB worse thanthe results achieved using the resonator(at 77 deg K) and theactive circuit(at room temperature) in separate packages. Thefollowing seems to be the possible reasons for the relatively poorperformance:

1. The radiation Q of the packaged resonator depends verystrongly on the material and dimensions of the package.The package used for the integrated oscillator is notoptimized for high Q.

2. Lower value of the loaded Q. The loaded Q in this case wasmuch lower compared to when used in a separate package.Asper the Quality factor measurements at STI, unloaded Qvalues of 3000 to 9000 were reported. The phase noise andfrequency stability performance of the integratedoscillator indicates a much lower Q performance. This Qreduction may also be due to the losses between theresonator and the active device as well as the lossy

interconnection between the superconductor resonator andthe active circuit.

A gold microstrip GaAs FET oscillator was built on aluminasubstrate for comparison purposes. As shown in figure 7 the phasenoise of the integrated superconductor oscillator is about 20 dB

better than the microstrip resonator GaAs FET oscillator. More workneeds to be done on improving the loaded quality factor of thesuperconductor resonators and optimization of package for betterperformance.

III

ICIL I

III. n 0

C4D > _

04 o

CCF) m

c- I

III

HARDWARE AS PART OF THE FINAL PROGRESS REPORT

I5 i. 10 stainless steel packages used for temperature testing.

2. 10 aluminum packages used for temperature testing.

3 3. One integrated 9.47 GHz oscillator

5 4. Two 9-11 GHz negative resistance substrates

I

II

IIIIiIiI

IIII APPENDIX 1

ITEMPERATURE CYCLE

ITEST RESULTS

IIIIIIIIIIII

5- E7' ' ' -EF- I =I E D -rE-=E;-T

OAVANTEKSTART DATE I CUSTOMER I QUALITY ACCPT

I 1I0

I ----- E5------------I -------------------------------COMPL DATE I TYPE OF TEST I AUANTEK I WO/CSO 'TEST COND.

I FINE LEAK TEST I JI-32 I___---------------------- 1Z2L.'. - -- ,

DEVICE TYPE I TEST REQUIREMENTS I BOMB PRESSURE ANDI MIL STD-883C METHOD 1014.8 I DURATION

QTY START I QTY COMPL I SENSITIVITY I 01-3

I /o0IXI0 X~' CClSECl

-----------------------------------------------------CC/SELOT No. LEAK RATE (CC/SEC) PASS/FAIL DWELL TIME3 SoL .. P-2,,,. '%, ,,-' ____ ____ ___ __,-,.___,_

003__0_ ___ /c - Feed .' e,) Au'o.

Ivl I

I _ _ _ _ e

cot ___ _ _ _ _ _ _ _ __$As_ _

AV-631/R-10-87

U I IN

(D 0lII 0 *0

* 0

5 IIN

U 0

(0

1 0 1

fla:

* 'CD

00

1 10

*0 ____

T.

ccA A

Cu

W4*0 0

I~ ;cu

* 1+

w 0

x La

- c 0 ----

0 10 0

*mm

S)' I C

-lo

cc 10

Kc-~

K--- +

A A

in c --7N

N 0 0

+ 1+

Vu I 10 _

U ' F

(0

30 r +

LL .

a: +

'C Cu

I

(I

I +

I100 CD

cu 0

M (Li0

W0

00 LO

0 +'0 '0

00 101-

3 O0 +

I z1 _ _ _ _Nm 0 C)2

LI _ I N

W 0U l 00

0 cu0< +

.,0 1

AAI I 1 I

Cu

U + +

wcco

mu

00

L) 0

0I N

LO 0

w 0D

0

Em 0 +<t

T 0 I-

A A

IIII

1 50 CyclesUIUIIIIIUIIIII

a E ; r~ I F= I 1-= n F STF! 0AVANTEK T ET

... -A-R-e- So cyc~l-sSTART DATE I CUSTOMER I QUALITY ACCPT IiELj

I I AC~CEP7

COMPL DATE I TYPE OF TEST I AUANTEK I WO/CSO TEST CON

I FINE LEAK TEST 1 31-32 1 Z6 " ---

DEVICE TYPE I TEST REQUIREMENTS I BOMB PRESSURE ANDL P-200o I MIL STD-883C METHOD 1014.8 1 DURATIONQTY START I QTY COMPL I SENSITIVITY 1 1.3

-- I s' °" -7 CC/SEC 3

( LOT No. LEAK RATE (CC/SEC) PASS/FAIL DWELL TIME

"1..0!2 , __,_______________,____.__

004 _____________ _I °°____ F4v M,

i " Oe ____ ______________

oo_ _ _ _ _ _' 4L_ _

005 Z( --,o

Dp0 lOJQo- _______ k-4k 0-e. M'I,k o, 124J Ag

I p,€ O. o- co,/

U SCA 100 ao

005

I _

~AV-631IR-10-87

(UDU 0) 0

0)0 _ ___ __ _ 00

I + r+

4c ILNI Iw0

MV

0i

ci) (U

I

U 0 +

w 0) I r~%0

00

0 c00 0

Vi. . C

*A A

CU

!

I

I <:

INN- 030CD

coo 0'O 0

I + +

NLa0U

I DN

0CU 00 - _

*0 (0

z +

w4 10

cc 0LL 0

* 0

I: 0 C

-) VI0 I

IA A

U cu

II 3"-0 CD

loo 00cu

I <II

W 0

MID

- .0 3___ __ _ _ _ _

- l u* CD*0

jU __ __ __ 0m 0

o (0

mm Z

coo EnEto 0

~1 CD3 L 0Cc 0

x 0

NI 0 - CC

I3 0

I o

I

(~0 0

N

mo 0

V'II + !+

I <Li..w 0C c-

I Cl),

x AnI m

X90 - ED_

vi. 0m I(u 0

o 00- _

* m

1 u 0m 0

rLo

A A0* CE

0 vc

mI

I mmV0D

00 '0M- 0

I + +

< <LI ILw 0

.0

Ir u 0oID 0

0(0

0 EDmm Z+

* coo0

III L 0w 0I E 00(

0 +

< 0 _ _ _

aa110 C,

_

A A

cu

0

3 + vi____ _ _ _

wi 0

I u 0mu0

ID0

0 t

mm+

coo

IIfIC)D

* LI__ __ __ _ ~0w 0 __ _ __ _ _

0 H

Ao A

0, in E

Is

Ii N

"~0 ED_ _

I~ 0r 0

I ++

IC

0 70

000I) 0)

0C, __) ____

000

Urn 0 +

II q-.

To 10Al AI 0U

0Ic

Inmc

0

TI 0

cu1+ +

w0

IiN"E

I i 0

00t

00 1n

ILL 0

cc 00 __u

V I. aa

AA

I (u

LIIq

'Do

I'%II+ +

< LL I

I 'Dm

'u

mm

N

LL 1

Cc 00~0

:0a

3 A A

cu

INmr%-0 0

Uq 0 0

1++

ILx 0n

II I

I _ _ _ _ N

mcu 0

(0

0 00

IdI0 t0

00 xcu 0

coU

1' 0~

A A

0 +U

.1 0

IIIII

20 CyclesIIIIIIIIII£III

E:NAmLU ALL -rGE~ I F= IE:n--EE3-OAVANTEK ~~A(CKr JoQ'Lde Dib-

START DATE I CbSTOtIERI QULT CP-'516Y ~#FH A) ;X0

COMPL DATE I TYPE OF TEST I AVANTEK I WO/CSO TS OD- I FINE LEAK TEST I JI-32 I

--LQ rJ ~ -- ---I -- --- --- --- ---- --- --- --- --- A -DEVICE TYPE I TEST REQUIREMENTS I BOMB PRESSURE AND

I;- & MIL STD-883C METHOD 1014.8 1 DURATION

-QTY START I QTY COMPL I SENSIT VITY OC~Pf44 4 rx

I,--- ------ ---- ------------------£S/N LOT No. LEAK RATE CCC/SEC) PASS/FAIL DWELL TIME

__ __ _ 3 X lp- __ _ _ __ _ _

rv _ _ flx o)

I ~ ~ ~9 4__ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _

I ~01 q___ _ I_ _ _ _ _ _ y_ _ In_ __--1_ __

AV-632. R-1O- ____

LI

C!

cu

+

I <LL IL

I I __ ___U)

NN

I1W 0

0L) 0

(DI +

I 10C z -g"

N) x

IrI

0____0

0

co 0 +

mo u

A A

I Io100 C0

+

w 0

1 .. 0__ __ _ __ __ _I

Tcu 0

0(0

0 +Inn Cr

ma 0

00

0 (Urn 0 +

a-UCX- u

A A

(Uj

+m +

I in

00C* d 0

0 +.w 0

NO N

Icu 00

mm 0

<1 0 I N

'I-'CcI _ _ CA 0

I o co N

ED00 0

* 0

3 1+ +

<I a-w 0

MID N

0 ___ C

(U. 0Tcu 0

0(0

0 +3 mm 0

ED~LL I 0

a: 0Cl 0

0 0U

E m0 +

<0 0-U vi (ii

I I H

CDM ctC

0)0 0

+ +~

w 0I _ _ _ ___c

CD T)

0.0M cu 00 0

Lo tom 0 +-

00 0I ctDoo Ny

i I -I

w 010

0 c0

0 U(

I A A_-

Vu

INC-1D

w 0

II

.0

I ) +

(D o

I0 _+

(D 0

I L__ _ _ _ 0a: 0

0

0Cm 0 +

n- II

(J*4A A U

cu

I Nw40 0I D w0 ___ 0

0

I + +

Cc.

I~uin

* 1 00

(D

z cc00 _ _ _ __ _U

* A

ED~ C

INI o

I ___ ___ __ 0I0cu

1+

* w 0

* .. o T___ _

* u (- 0T u 00C _ ___

0

0 +03 m 0+

Io (0__

(DC

U- 00 Cu

cc1 0~I

A Au

*L Cu

m131

U u0

1+ +

w0

cu. 0m)cu 00

I (D0 +

(D NN

ai:

1 1 'N

(DEDtoo0 0___ _ __ _ __

III L

NN'D13

00

m m 0

00 cCD '.4zl

LL 0

cc 0

c~ 0

Al I*) .. U

I q CLC

UUII

1 10 Cycles

IIIIUIIIIIIUII

-EFRT -I IF1 I E-= n TrFE ST

O AVANTEK A4- /0 f ycAT

START DATE I 'CUSTOMER I QUALITY ACCPT

I COMPL DATE I TYPE OF TEST I AVANTEK I 4GSQ-A,9l Akr *'TEST COND.I FINE LEAK TEST I JI-32 I

I I3 DEVICE TYPE I TEST REQUIREMENTS I BOMB PRESSURE AND

I MIL STD-883C METHOD 1014.8 1 DURATION---------------- I ------------------ I

QTY START I QTY COMPL I SENSIT VITY I -r T-

/0 1 I CC/SEC

/LOT No. LEAK RATE (CC/SEC) PASS/FAIL DWiELL TIMEez , ; ... 4 ____.c_

Iz-o.- a Y---, ___-___

0 ,5 I o'- q_9_0 y in-If

voI

&2 T1' ____l " I nW09K vd AxbtA

I _

I .__ _ _ __ _ _ _ __ _ _ _

AV -631 /R -10-87

I LLCc 0-

0

co +

Wu 0

1 00 xz CD

ILL 0w 0

05 0

mo +

III 0 N F

ccc0

Mm N

(DED 0 0

j ++

vi 0

m

00 y

00

I 00

0. N-0 0IC3U UU)

I 00I

Imm N(D

Mo 00 0

Cu

< <LLCc 0

En

cu 0~jo __

0

LO1 mm +

00

II 0 N0T

CDI L z 0Ld _ _ _ _ _ 0

03 00 +u

T-0 F-__ _ ____ En__ __ _ __ _U

A A

(r) 0 0cu 0 _ _ __ __ __ __0

0

w 0

MID-I __ _ _ _U

IcU 0

1700

cI 0

0 N:I _ _ _ CDLL z 0

___ "I,_r_ 0Cl ul0

00 cu

m_ _ _ __ __ __ _

x 0I

I cu

Mo 0

Ii

I + +~wi 0a: I

MV

cu - 0mcu 0

0

tomm +

Ic, 0 N-170

00

00

I _ __

A A

I cu

U mm N

* 'o 0*v 0

cu3 + +

w 0

cu mn N....

MWu 0L) 0

0

mm+

0TCD

LL Z0I Li _ _ _ __ _ 0ITT 0

I 0

0 cu

C"la)FEjj~ A

I ('A

Ir

* NCD

* .. 0u~cu

I + LLi

w 0

CDID

00Mcu 0

0(0

(D 000 x

0N0

1-1 Z- 0

II 0

m 0_ +CU

I 0 H-

AA

INT

0

I +L.w0

I

Icu

100

LO 0ICD 0 M:

LO0

1 0

<*0

x 0-

0 +0I

I cua .

CDcu 0 0( DO 0

cu3 + +

< <LL 11w 0

I _ _ _ _ _ _ __ _ _ U)

Cu* 0Icu 0C., 0

(0

-(Doy

I 00

3 LL Z 0w N_ __ _ _ _ 0

0

I K. CuM 0-

U <h Cu

I++fiin

. .D

w 0

cm

CO 000

1 0 00(

LL +

I l N

II0 0

m 0 +

CD

A A

I (U

1 1:N 0 0___ _ _ __ _

* + +

I <LL

m

m IcuU0

(00

"0L 0) IzD 0

0-4 +coU

I -IN

AL 0IL CCl ui 0

x 0II r

IIIII

5 Cycles

IIIIIIIIIIIII

i Nnaiuu QERTIFIED TESTOiAVANTEK Af -5 yleg ni-rA 2,

START DATE I CUSTOMER I QUALITY ACCPT ACCEPI

---------- --- -------- - 04-- - - - - - - - - -- - - - - - - - - - - -COMPL DATE I TYPE OF TES I AUANTEK I WO/CSO 'TEST COND.

I FINE LEAK TEST I JI-32 / 1Z' I ' /. .. .I ---- - - - - - - - - - - - - - - -I -- - - -- - -- . .. ... . ..

DEVICE TYPE I TEST REQUIREMENTS I BOMB PRESSURE ANDI MIL STD-883C METHOD 1014.8 I DURATIONI-------- I ------------------ I

QTY START I QTY CMPL I SENSITIVITY I 6/ ,

10 Jc/ 1 po~ CC/SECI s,-OT No. LEAK RATE (CC/SEC) P PASS/FAIL DWELL TIME

00 Z I IV/0-1_

007 1 X,*"- 00

oo-/7¢'

I _1_ _ _ _ _ _

ool 714- 1

00 zyo -,

0o7I'/ioi__ _ _ _ _ _ _

D67 __ __ __ __ _ __ __ __ _

I _ oo; l~oo"77 Vto"

I __

AV-631/R-10-87

INDtoo0

0

g + +

IL III

IEE

0 _m

u.,0

I N_-0

X

& m 0 +

10 %

.0 11 (

- I1C

: m Nr-0 0

u~1 (U+ V4__ _ _ _+

< LLI 0

m

I u CDICU 0

to) 00y

I mm 0I (DO

w 1

x I0

0 +

I Ia:

-T 0A A

I (U

I r fin N

EDMO 0

0

5 + +

w 0ccHI x E_ _ n

I Iy

0 a0

0

+

nfll

ItooN

0 N0m

Lw z _1 0EC 0

I m 0 +cu

C~) H30LI) _ _ _

A A

t cu

I +m<i IL('o C

I w 011 1 0

(0

mm +

IT 0

10 'TJ I

CI 0 i 10

0~ - u_

(n 0

A A

Uc

£ mm N

a~o 0NO 0

C!

IIw0

I _ _ _ _ _ _ _ _ l

K2 I(!U* 0m) Icu 0o 0

I +

(DoCD

I 1 N* 0

LL z 0

w N" 10.

Cuu0 H

A A

ru

INLI

cu 0 01

w~ 0

10131

I.U - 0__

mo N

* 0m

'Din

N _ _ _0-

L Ii C)Nw CD

cr 0r

0 Hu

IIc

O1:00

I<L.Lw 0

cc-

I EcnMIDCu. CD

cu 00

+

*mm (0

- 0

A A0

NN 0 j c

I1 1

£0

I

-d- CD

0

I Y Cu

LL 0

w 0

0 +

A A0cu+

I0

+ +

NN..

cu CD

Ii~~ 0 I.

Ao 0

II1L ______ _________ _____x__________

mm0

finvi+ +

<'J C0* cL _ 0

I .. w 0

m

610 ID

mo 0

m Iu< +D

0 0-

LJ .. 0 I in

A A

I Nu

'II

I

I

£I 0 Cycles

IA

II

I

I

l 0AVANTEK E -AT V Ej' START DATE I CUSTOMlER I QUALITY ACCPT

COMPL DATE I TYPE OF TEST I AUANTEK I WO/CSO 'TEST COND.

c I FINE LEAK TEST I 31-32 I I 'I--------------------------------- q .....

DEVICE TYPE I TEST REQUIREMENTS I BOMB PRESSURE ANDI MIL STD-883C METHOD 1014.8 I DURATION

QTY START I QTY COMPL I SENSITIVITY I 7zk5 s.,5

I--------------CC/ SEC( LOT No. LEAK RATE (CC/SEC) PASS/FAIL DWELL TIME£AI CA.. J-i

____ ____< iTo~ "_ _ _ __ _ _ __ _ _ _C.? < £Po =__

03 _< _ _ _ __-_ _ _o44 j __ ___

<Z io 7 _ _ _ _ _ _ _ _ _ _ _ _

KO " -____"

I ,__ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

0 I

I __

* _ __ _ _ __ ____

!I

II

w 0TI

"N.umT

S 1+ +

ft c-0 0 y_ _ _ _ _ _ _ _

I0,to L)

< 1

YE D

Icro

T)I

I .

S mm NTV

* 00 C!

LL (I

M'13II u +

I 0ED

mn N

"0 0

LL 0w 0

0

+

A AmVtu

m

100 CD

E O 0 0cu

I 1+ +

L L ILW 0

II

0 ED

T-, 0

co_ 03

CD(0

vi Um!Im -

flyu

0

A Al

Cuj

o wm N

l Wl 0cu

I 1+ +

I L a-w 0

NNEED

Icu 00

(0

m Hn

It N

LL 0 CDIL 0

0 Cu0 +

0 H-

U) c. I

3 Cu

ICA((

(DO 0cu3 0

(U

1+ +

ILLw 0

I-)

mn Nt .. 0 __ __ __ __

Cu Dcu 0

0(0

mm +

0 - _

00

0 0q N 0 0Uw 0

0-0

N-1 0 cu0 0H

1) HAnA A

a (Ui

mo (DM 0o 0N* __ 0

>+ +

w 0

MID.. 0 CD

Iu - 000

mm +

a

0I

0 - _

N1 0 0u

(U 0.. 0

T. (UF

AA

NO __ 0_

0

3)0 __ _ __ _

++

5 0mfiEn

NN

T- 0i1 0

~~LOa I-

A0 A

3 ~=1 C

.1u

mm N

00

cu

I ~LLw 0

0 . co

0

C~)C!

LO~~ 0u

m 0(X

10000I

I--

:E 0 0 C

0h to

A A

I cu

LIu~cu

LLJ CL

wo

x i

.. 0m

L) 0(0

mm+

0LL 0L

CCu

000-

mj 0 +

A Au0 +u

I F

IN

'"'01q4 0

I ++

U--

(131 N

A 0

nL) 0I __ __ _ ___ ___

I f 00-

N

CD

w 0 0

0 +N

V 0-

I afA A

cu

II

S mm Na0 0:

0

u 0u

1+ 0

w 0o

E.~ *.0CD(I IL 0cc 0

0

8 mm +

Q.0

AI 0

V0 Cu

IN

c 0

0

w 0

om+

NICD

LL 0

(0

2:0:

i I IT I iNAL A

LU _ _ _ _ _ _ _ _

I m I(D*ma 0

Cuo 0

1+ c+

I L 1w 0

mno N

(u 0

00

m m

'5 . _ __ __

I CDLi. 0

Cl 0x 0

Ou

Ao +

A A

K' cu

I0

000

I L a-Wd 0

Cl F-

00C., 0 Mn 0m cu 0

(0

1 mm +

mo

II NI)

IL 0a __ __ _ ___ _______0

0INI c0

A A:10 I-

I mm N

* O 00 000 0

c'J.- I

I + +

LLi 0-I L 0r

MA U N

cu. 0ICU 0

0

mm+On Ci:

3mCTY

LL 0cc: 0

0

Cr

A _A

I-.'I cu

mm N

VO 0

I 0cu5 1+ +

IL 0-w 0cc I-

mmn N

ED"N 0

(00

* mm-~ nn a

(D

w __ _ _ _ _ 0Cl 0

0

m +

A A)

.3 cu

Iu

I mm N

LOo0 00

cu

1+ +

w I

I mT

* mm

I m +

~~ LL0wi m_ iii_ _

a: 00

0~ cu_ _ _ +

> A:

A Au

q-1d

mm co N

CD 0 0

0

1+ +

W 0

m

.. DC

0~U

u

1 o m

NO

N _

I I

ClC0

M : 010

0 >i

i;4-< 0 a:

A A

I cu

00 0!lqo 0

I-0

5 1+ +

wq 0

(3 rr )i 0cc 0

mm +

U c

A AL

3 Cu

U NC,0I ~ .... 0

< cu

I 1+ +

11 0

cu 0

ED

1 mm +

a

CD

Cl 0CiI 0

m _.1 &-% +

U T-II Cu

I

!AV~F~1-= M~4 -I-C ~E~ I F= I 1- D -F -=13-TOAVANTEK ?bsATA np% 1 42%

START DATE I CUSTOMER I QUALITY ACCPT LL

1,- - - -- - - - - I - - - - - - - - - - - - - -Ir-I

COMPL DATE I TYPE OF TEST' I AVANTEK I WO/CSO .-. , 'TEST COND.I c~j~q I FINE LEAK TEST I JI-32I----- ----- ---- ----- ----- -- -, - --,-- ,_ .,_ . ... ._- __.DEVICE TYPE I TEST REQUIREMENTS I BOMB PRESSURE AND

I MIL STD-883C METHOD 1014.8 I DURATION_I. _ --- 1 - I -

QTY START I QTY COMPL I SENSITIVITY , 2'A...... ) ] 1 Z4o " 7 CC/SEC-

_ -------------- I------------- --------------------------- ----()LOT No. LEAK RATE (CC/SEC) PASS/FAIL DWELL TIME

i~~~ ~~ Ol Ito " -b $ o .IA i< 110 " 5

'e.Z&d -,,

0$1 < >(ko 7 ~- _ _ _I -2 7do -'7

I _ _ _ _ _ _ _ _ _ _ _ _ __ _

II '__ _ _ _ _ _ _ _ _ _ _ _ _

I ______ _______

I AV-6311R-10-87

IIII

I 75 CyclesIIIIIIIIIIIIII

cE= rT I F I L3- FE-I AVANTEK 775F c F- ,5= -- S

: UI % ~ 75 c,(cles AT

3 START DATE I CUSTOMER I QUALITY ACCPT oc

S- 4ccc4....---_ _ --- -- -- -- - - - - -- - - - - -- -- -- - -- -- -- ----COMPL DATE I TYPE OF TEST I AVANTEK I WO/CSO 'TEST CND

I FINE LEAK TEST I JI-32------ Z------------------- 1 Z L~~

DEVICE TYPE I TEST REQUIREMENTS I BOMB PRESSURE ANDI MIL STD-883C METHOD 1014.8 I DURATION

QTY START I QTY COMPL I SENSITIVITY 1 .

,l f . I S i l o " C C / S E C I

(9LOT No. LEAK RATE (CC/SEC) PASS/FAIL DWELL TIME

ck

II

I~

I AV-631/ -R-lo -87r e a... i,/, .

00-7/0r- L

0-, y O

/00, OAS

cwl

AV-631/R-10-87 °; ..........

IN

(DO U)C() 0 __0

3 + 1+

I LL0.w 0

I o 10 N

ECD

Ic 0

I C!

03M +

I 0 TO z:(DO y

0 0 _ __ __ _ _

w 0a:_ __ __ _ 0

00 c0

co _ +

F-

[(nUL) --TI

AU

II

I O I(D40 I 0r.%o _ _ _ _ _ _ _ _ __ _

Icy 0

I+ +

I CLWl 0ccNNII N

(D

2:0 0"o 0ON, 0

O0J o

cu to

0 +

-Z cc

00 T

II I IoLu 0

I N

m 00010

A A

ICCM 1

1)Tii

coN

I J 0

* + vi

III IL

- N

cc.

Xo 0

ON 000

XYO z _ _ _C___ _c

III IN(u CD

LL 0IL 0 0

0 cu

0 +

T-0 H

A A

cu

III N

CcD

(U 0

9. 0

wm 0

(UO -D

IL 10

I 0 I0

;* II[

A .0

I (U

Ir

w0

1 +

I <a.

Eo V

cu -1 c 000

mm. 0

00 0

I<D 10cr 00

100

A A

cu

IN

I ED

1+ 5 +

w 0cr

x in

MID N

mo 2m

IF 0(U (D

IIu 0

NN

A A0

0 +UfI

00 0

4o I. CL(vLo 0

Ccu

03: I

1W 0rN000

£~ 01

oo U)

0: L)0__ _

0

0c

IN

cu

00cu - 0

_ _ __0_ _ _ U

(~ito

0o

NOU mo U) I

I I

ii I

LID

g L 0'0 '

* o 0

A A

I cu

mu o ED__ _ _ _ _

m 0 0

1+ + _ _ _

aa.

I ld0)

(0

lz a:-mo

IL 1

00

A Au

VA CuI c

14 0

I + vi

<Li.(

I0IFF

0 _ _ ID__ U

0u

I~cu

IL I) IK-1

<V

.1 cu

IIIaI1 50 Cycles

I£I0£aI0ISIII

I OAVANTEK ,,ST

START DATE I CUSTOOMER I QUALITY ACCPT CL... o - -------- ------ ---- ------ ------------ ----r J - - - .- .-

I COMPL DATE I TYPE OF TEST I AVANTEK I WO/CSO 'TEST COND.I FINE LEAK TEST I J1-32I ----. S.e

------------------------------- ----- ll~s------DEVICE TYPE I TEST REQUIREMENTS I BOMB PRESSURE ANDZ. L 5;-2ccc I MIL STD-883C METHOD 1014.8 I DURATION

c- --- ------------------ I

QTY START I QTY COMPL I SENSITIVITY I

(3s LOT No. LEAK RATE (CC/SEC) PASS/FAIL DWELL TIME

oZ0017

I ,L. __ e°Y.a5j(00Z

oo0 J ,

0?

iI l

AV-631/R-10-87

I0C--

LL CLw 0

MT)

cu (D

W) 0

1m

cu 0 uL) 0

W IL 10

0)0

I4l) 0

A A

rd

Ic

V*000 I0

C!

3 1+ +

w 0

II

0.0 c0

ci 0

0(

mm +

V-4~0 I

I D

LL 0

F~cu0 +< H0e,

F-

I CCToi inc

mm

I +

w > 0

(U

I 4L m

MIN

.0 m_ _

4 IC~j 0C- 0

00 (

0 to

crio

U4-

I 0D

0L 0 0-l 0

N CU

I AA

Io

0 6 0

i1 + +

< ILw0 II-n

I .00__ CDcu. 0X(u 0

0 00

0 EDI +Im

I0

_________ ____ 00 Ivi0 +u1 _ __ _

Oo1-a A Acu

3 03 1:(DO I Dcoo 0

* 0C!

+

I (L (Iwc

Sco

00 0

0 to

* 10 Hl* cc., yl _ _ _C

II

Inmc

to (DO

I~ + +u

< LL

I Ec 1

0Ccu -..................

£o N

wwI L 0I1

lon~lo cI +

IA A

I ~1 (u

V0 I 0

+ 0

I 0L

w 0

MI N0.0 _1_-0 3__

(U Lu.3CU 0

00£O 0 0~

0

0 L L0

LI 0Er 0

m 0 +

'ri. Li

A A n

cu

"~0 Lo

TIo 0

* cuI + vi

I<LL. s.+w 01

Urnn60.0m cu 00 0

0) 0o (0

Imm

z IT

0)0 0

0)0

0

m +

m1 20 F-__ (in3A A

I TI cu

*n ~0 -. 00_ 0

w 1cc I-

0u CDiu 0

00

0 (0

1100

LO 0

w ___ u I 0Scc K10_

0i +:

IL CC3: 0 F-

IS3SI

20 CYCLESIIIIIIIIIIII

c, kF-e. F;ER -f F I E D Tr1EST-

OAVANTEK I 4-A

START DATE CUStOMER QUALITY ACCPT "C7

J -a- -, ... ..-T_COMPL DATE I TYPE OF TEST I AVANTEK I WO/CSO TEST COND.

I FINE LEAK TEST I JI-32 I A-I----------------------- --------

I DEVICE TYPE TEST REQUIREMENTS I BOMB PRESSURE AND

-- 00- 100 MIL STD-883C METHOD 1014.8 1 DURATION4 --------------- ---------------- I------------------IQTY START I QTY COMPL I SENSITIVITY 6- .ek..3

( )I - CC/SEC ----------- ----£ S/N LOT No. LEAK RATE (CC/SEC) PASS/FAIL DWELL TIME

too.&I _ 007

I __ __ __ __ _ _ __ _ _

I __"_

I ".,,_ _ _ _ _ _ _ _

AV-631/R-10-87

I mm NCD

NO o

LLJ 0

w.U 0

m

1W 00

0 > 0M 00

00 c00 Cf

CCC0

x 0

0 0uM 0 +

<0t'

I 00 0A A

I cu

II

, (' D(no 0

I 0

+LL a.w 0cc F-UK)

Iu 0

X N 0

- I

41W 00

0 +

I A D

00

Iq 00n

LL 0

3 02 0

00 ___

J0

A A

cu

COMD

vlo 0P! 0

5 + r+

wU 0II

3 .0 CD(u. 0

00

m (D0m +

Ez All a:

00 En

IICD

LL 0I w_ _ _ _ 0cc 0

x 0

U ' 0 +U

m- 0 _ _

A H

I Cu

V4O CDj

0ccU

5 Wu 00

ECD

0 0(3) 0

10 0

(0

0 +

in U)

A

III I N

INDrmo __ _0

Iil 0C!

5 + +

< <LL ILz 0

co:

0~ CDN. 0

- 0 00

0 +0I m +

I (D

I I N

LL CDI _ _ _ _ 0W_ _ _ _ __ _ _ _ _ 0

Cc 00

IN, cuI +

-0-NJ__ __ _ CC_(

A- 0AF

U W4 Eni

m co__

mIII

nfLL II L 0

cc

cu (D

mcrUI K.0In0

mm 0

mmi 0 En

II I NCD

LL 0U L 00(In 0

x 0-

0 c0) +_ _ _ _ _ _ _

a A Acu

* mm N

C!

.4 0

MID

(U 0

0 +

(U.U09 w 0

Cl 0

0 c0

0 +

Xo uC-) _ _ _ __ __ _ _

II I N

TI I

00 0*6 0

+ Cu

< LLw 0

Iu 0cu 0

o 0o0 0

I CD

II NI l 0x 0

N, 0 C

I 0-(I CC

2:0 :

I A A

I cu

I Cu _

(DO 7 )0cu 0 0

cu3 1+ +

I <Li (Iw 0cc F

M1_ _ _ _ __ _ 3

1W 0

0 0

mm 0

0 I KI +~t ~

.0,

I . 00J in in _ -_

LI

vi0 0

cu1 0

U w 0

E l +__ __ _0

tw 0~ NN-

II

wC 0U Co _0__

0

M 0 +

Cl) I

U" n NO

'-4 I

IIUII

10 Cycles

IIIUIIIIIIIII

IOAVANTEK Ae f cyleE T DE

START DATE I 'CUSTOMER QUHLITY ACCPT JACCE .,

I COMPL DATE I TYPE OF TEST I AUANTEK TEST COND.I - FINE LEAK TEST I 31-32 _

* -- --- I-----------------------------------I DEVICE TYPE I TEST REQUIREMENTS I BOMB PRESSURE ANDI MIL STD-883C METHOD 1014.8 I DURATION

I - --- --- --- --I ------------------ I3 Q-Y--START I QTY COMPL I SENSITIUITY I I- A

--- - - - I (/ $ CC/SEC

3 S/N LOT No. LEAK RATE (CCISEC) PASS/FAIL DWELL TIME

I I --- / X /!o-- 1__ _ .2n. .-I x ,I - ___-__

x 10

x 10./.-Cf2 I 0_---_

I~ ~ 0j I_____ ________ 10_---I

AV-631/R-10-87

'DI

Cu 0 LID!

* Cu+ +

U 0

010

'Cu~

II1000

Cmot

L ZiC!

0 1u

0+

00

X 0 I-

__- -- -9--]U')_AA

vi

I-u

Nuo 0D

< ILUciw -1- __

.. 0v____iI

IWO _0 _ _ -

2 00

rn ca

10 0

* 0~ ~ L __ ____Cl

U )v AA

I ~ ~ 1 u

IcKNII I

11+1 ,10Cc---I ~~*~

cu I

I m -- __ __

Cu. I 1too -

* mmco- ,

A AD

* Cu

I6

f

M.-; 0

+ ++

w 0F-U t)

MT3 NI:

0 CDcu. - 0

Cu

Ir

I 0

0 -- .-

I Cu

I,+

3 V0 0ma0

1++

I wL

U m)

'0

I o10

w(* +

* a _ _

.0

AL ZiIl : 0

U1

jDr

11

*i 0

m 0

-T

Cu.)

rcc(~10

.0;I -

tci 0

+m N

100 0

cu 0

+

1 0

my 0

M:. 010

0 c0

0 +

Aa A

ruN

INN- O 10 0

LO 0

('j

1+ +

wL 0cc I

In. 00

o0

0001

~-I0

U CD:E 0

__"_ ____ ___ _

A0

IIr0

-mm N

U (uo 0I 0

0cu

1+ +

U <L1 Cw 0

T- ly

L) 0

I +mm ---

ILO 00 I-

wJ NZ 0

0

0 cu* 0 +

I Mo C. _ _ _ __

A A

cuD 0

w 0ccU

0(-'U- (.

(U. 0

"003 O 0

m

II N0 I:

LDLL z 0

_ _ _ 0a: 0

0 +U

U) A A

cu

IIIII

5 CyclesIIIII£IUIIIIIU

0AVANTEK A r ASTART DATE I CUSTOMER I QUALITY ACCPT

------ -I--------------------------

COMPL DATE I TYPE OF TESTY" AUANTEK I W0/CSO 'TEST COND.I FINE LEAK TEST I JI-32 I I . /4---I I------------------- ---------

DEVICE TYPE I TEST REQUIREMENTS I BOMB PRESSURE ANDI MIL STD-883C METHOD 1014.8 1 DURATION5c p- /'2,0 P-i

3 QTY START I QTY CMPL I SENSITI ECTY I

10 /0_ CCSECI (S/ OT No. LEAK RATE (CC/SEC) PASS/FAIL DWELL TIME

, / Sc r2, 7 c, l0 "11 .<

ooZy/

0" 1 L 17-Z 9-o04 7So "I07 74," q q

00-~o q_____________I_ _I

0447,60~

I _

I _i

AV-631/R-10-87

LIItI

V0 I I

cu3 1+ +

Il 0a: I-I _ _ _ _ _ _ _ _ U)

(U(D

I c'o ocoo

I I N

_- _ 0__ _ M_ _ 111_ 0

a: 0I'--E ___ (u

m _ _ _ _ +

U . U wl . w 0 :iD F

A A U

I cu

INNCDO 000

cu

I +w 0

I U)

1W 0

U)l 00 Iq 0

0 N0

cc 0

0cr 0II 0

* _ +

0-~4 LUC0II.% Ia Cc, U -AA

I Cud

00 01 !_ _ __ 00

cu5 + +

MID0

cu - 0

CD00.

00

000z LO

E -m

i0 NuM 0 +

CDTo. Z

A A

a ii

I Mm N

ED0

I r. 0

cu1+ +

I <LLw1 0cI

MI

cu. 02:cu 0

(0

1 mm +1 0 T) cc00 l

0 N0 m

z ED

LL 0~~~ 0It 00 cu0 +

cui

UNfl~0 0

0

cu

+ I+

Lw0I __ ___ _Iin

CU 0U-, 0(0I +

MM

- CC cc

I(0I I

00

Ii 0

I m 0

CL cu

AA - I

* Cu

I v0

+ +

< <LLMw0I I__ _ _ _ _ _cc__ _

Cu.

C,.00

MCU 0

00000 I

to 00000

' L z 0DLTL 0

I 0a: 0 n

- m 0 N

0 +

3 10 UHFU''A A 0

I cu

Ic

I 1+ +

w0

'NN

(u. 0m cu 0

) _ ___0

(0

+

'o 0

LO 0

0 I- CD

LL z 0

0i 1 0m 0 0

'N Cu

INVic 0

(DOU1+ +

I <ii~W0RHI I _ _ _ _ _ _ _ _ _ _ _ _ vi

+ + ,____________

cu. CLw 0

MID(N

00 0

0 N

000

00

cc 00 ruI m 0 +

a- Cl

A A

CU

I m Nco 0

cu3 1+ +

< <LL CLw 0Cc

cu. CD0

(Uu 00

I +I a

Ii I N0M

LL Z 0LLJ 0

x 00

A Au

Ic

- mm00 CD00 0

I (U

I +

mIN

00

C', 0to

C1 m

1) I II

00

0 (um 0 +

mI H

A A

I cu

UIUI

I 0 Cycles

IIIUIIIIIIIUIU

tt

w 0Cl l"

w 0:

T- 0

mmcn CD

0- 0Y

1- mm Z

0 cum0 +_ _

u 0-

A A

(U

I mm NUnD

00 0DN 0

ru1+ T4

ILL (Lw 0

ft in-

MI1 N_ _ _ _ __ _U

-Cu 0

0

m m Z +ELI C

U70 M

N

LL cn0w 0___ _ _ __ - _ 0

01 00w

m 0 I_ _+

m10HI En; A__ __ _ _ _ __ _ U)A A

I vi

U N

- (Do CD-ma 0

'-I 0

+ 1+

ILLw 0

_ _ _ _U)

- U 0m cu 0L) 0

m010 (D+Imm Z -

ma* Y 0-0_ _ _

I - ____

LL QW 0_ _ _ 0

El LL0x 0

N. cu

0 +. I

Cul3 U TI lf c1

3N

- mo CD* o 0

0

cu

I + +

ILLW 0III I-

Uy 0

Li 000

0 +0

a I

N_ _ _ _ _ _

LL 03 L 00

N. 0 cuI0 ~___ __ _ +

0HCLIU I-

3A A

Cu

U o co N

00 0D

0

cu3 + +

w 0

IO ED

(0

u ) 0_ _ _

0D I

(0+

W 0

00

w 0U

3 00 H

aA Aw'

CuD

"0I 0* o 0

3 + +

ILLCLiJ 0

I . mn(NJ. 0Icu 0

to) 0

Po +

Iq 0

X N

LL In 0Ld 0 _ __ _ __ 0

Cl 00

0 (01m _ _ _ +

Ao A

A AU

I.u

Umm N7:CDIwo 0

_ _ _ __ _ _ _ _ _ _ 0

0

Cu

m

~0 ED_

u 00

0 (0

ma

-j-

U- '

U) 0

00m 0 +

'N Cu

01 U-U) ___ _ U)

A A

U 0300

mo 00w 0

cu

I 1+ _ _ _ _ _ __ __ _ _ _ +

ILLw 0

II

0

CU. 0I cu 0

L) 003 0

0-LO

U) _ _ __ _

U 0 _ _M: N

< TI

0 F0

m0 0

UV1 U)FA A

I cu

N1 N

Uqu 0

w0

U- cuu ___________

m

CD 0Dcu 0

w 0 00

0300 +D

.. 0

CCDU Cl)_ _ _

Ic

3 Y 0

cu1+ +

IL a-w 0

WVD Nm . .0 L__ID

m 'u. 01W 0

00 (D3 mm +

m 0ma0

(0 0

U)

U- 0D

a: 00M: 0

U T-4 in_ I__ 1 __ __ _ _ U)

A A

I, TIcu

_DAVANTEK ENVIRONMENTAL SCREENING FORM

STA-BAKE DURATICN: _ _ _ REF.#: LOT#. -

START DATE:__ STARTE TEMP(C)*

COMPL. DATE CMPLTIME TEMP(C), TEDLJ

COMMENTS:

TEMP-SHOCK LO.TBAP: SAJTI ME__ TRASTW LOTw- LOT ___

TEMP-CYCLE ITEMP:- so,,TUn CYCLEST

START DATE STARTTIME _ OVEN _ Mot

COMPLr_ DATE COMPL TIME _ REFA#:_ _ _EC _

COMMENTS:ACCELERATION/ DATE _ TEMP_ RALS.- ACCEPT- LOT#.CENTRIFUGE TIME ,AXIS - RPM - RE.ECT - TED_

COMMENTS: REF.#._

VIBRATION STARTDATE _ COMPL DATE. LOT#.

COMMENTS: REF.__

LEAK-TEST FINE __ _OTH - LOT#t.__,PS SD707/ P__S P _.

START DATE. 9 -!0-91 STARTTIME 14'0 t/0 40 A CCEP

ComA.- DATE. - f Cc*PL 1M4ELgL /o, (( T' (0Y 3EEC ,Yo

COMMENTS:. 0/f -

LEAK-TEST SLEAK OTHE:__ LOT#:.PS I PS PS

START DATE: STARTTIME TEC__: ACCEPT

CCMPL DATE. COMPL TIME t REJECCOMMENTS:D BURN-IN ATtCN" T- - LOT# ____.

START DATE_ START TIME TEp I t TE__

COMPL DATE: COMPL TIME TEMP It _T

-OVE. VOLT#.* CURRENT#. - REF.#: It @ 25C:-COMMENTS:

m TEMP-CYCLE! LO.MP - SOAKTME_ TRANS TglE,_ LOT#-.

I POWER BURN-IN HI TEMP:_ SO KTE VLT. CYCLES

START DATE_ START TIME: TEMP I I_ TEHt

COMPL DATE. COMPL TIME:_ _ TP It_ TE/ :t__

OVEN#. VOLT#:- CURRENT#- REF.

COMMENTS:

APPLICATION REVISIONS

NW A.WSLY Um ON LTR c.. DESCRIProN DATE APPIoII D

320-557963 KU-BAND DSO -- 00 INITIAL RELEASE ,z -*M'C* , .D_,.,_ _

I I

~ Z LLt I 1 o~.SHEET -,44 454 I 0 5SHEETV - 4 54647484950 51152 53 54 55 56 57 58 59 60161 62 63 64

EE!28 30232 1 34 35 36 37 38 39 40 41 142 43

SHEET 6!6 20 281 22

SHEET 2 3 9 0 11 12 13 141 16 17 18 19

SDIUNSIOS AnCES ftw" L.VELASOEZ I19O VAN,TOLERANCES c_ __ _

!1_ _ _ _ _ _ _ _ _ _

,.--,,0O+ ,C'v.aL1 APIQ$., - -S

Ali ,S .AS , MFGLNG MECHANICAL OPERATION SHEETO ,EPJI*IS , I 010 , UL NG CASE ASSEMBLY ScP-/ o°,'/ 0o

Owl mouse Apto slala Ott siqm, A 24539'4- 0NTSCALE Tel'S jIWwING __ _ _ _ __ _ S E WE)GH ~F REV

MECHANICAL OPERATION SHEETPART #:32O-,Ss8/22.001 IREV -- -W.0. # 5 A c 120 S-PART NAME: CASE ASSY Scp-oo IO

START / FINISHSTEP OPERATION DESCRIPTION TREF DOC 1 OPER DATE I"TY jor

1. VERIFY BIN-UP 3 20,

2. SOLDERRF FT(2PLACES) jAP0286CL

________________________ IAOJ-557963 ~

3. VAPOR DEGREASE PARTS PS-023886 -/-

4 FINE LEAK CHECK PS-029238 , / /j" f-

5. VAPOR DEGREASE CASE PS-023886 £~2-c/

6. 100% MFG INSPECT AWS-

AD-557963

7. INSTALL RFIDC CONTACT (2 PLACES) 340-026489- '-' //00, t/ °47 / /AD-557963

8. EPOXY THRULNE INTO CASE AD557963 v/APS-505313 - \//A

9. JINSTALL RF FILTER AND SMA AD-557963 / '-( /. /'-

10. GAP WELD CONTACT RIBBON TO THRULINE PS-501206____________________AD*-557963 _____ - -

11. CL.EANCASE PS-023886 re ?//q/ /

12. 100% MFG INSPECT AWS- / /AD-557963

13. 100% QA INSPECT AD-557963JI-0235

SIZE CODE IDENT .4O

OAVANTEK A 24539 MOS-558/Z -

11-I-4SCALE IWEIGHT -- SHEETaOF QL REV

* mm NTY 2:I mo CD"0I 0

_____0

0

(U

I ++

w0I :I

- (U.CDO I(U0000

ED

00 a

II-7N

Li. CD

a: 01l 0U 0(

m __ +

A A

I c ~

'01 7vl E

'-4 ____ ____

* 0

01 1+ U)

*I' (D,

x cu 0* 00

3 (0GO

+

IYICD

* o 0y

U- 0_ _ _ __0_ _

5 A A

* '-u

Imm NVr)0 0

* 0

1+ +

w 0-

mmn a

0 EDjo _ _ _

0~0 0

V*'~LN

CDLL IN,0

00

-. 0 H-

L) _ _ _ _ _ _ _ U)A A

*) r4

I mm N

(DNuO 0

0

1+ +

I- LLwL 0

m

M Rl 0U) 0

0

LO 00

II

ILi 00

r 0

CC

IE .0 H-

A A

Io

IMC

w 0+ I

INN

ED

0~LI

In 0IT H

10

CU~

I' I 'III- IU

Imm N

"00 0

0Cu

£ 1+ +

I L L1w 0

fIn

U CDcd 0

L) 000

(D

mm >+

* CDIL 0wu 0

Cl 0

00

cuu

IC

N

0

14 0

£ 1+ +

S LL 0wd 0

m lMID N

0 0D

u 0(0

mm+

000

Er 00

Lfl +

A A c

I cu

Imm N

(Dmc 0IM. 0

1+ +

IJ LLw0

II

I'O

T u 0I ) 00

(0

fmm

00 2

I~ NI:Iv~L ILD

cum +

~0

A A

I ~ru

ININ

iI

(DO CD

0

cu

I+ +

w 0~0

'L

CC(U. 0

, -cu 0

UO 0(0

L~cD +

ED

N 0U

Em +

* El

A c u

I '

INNOlo CDWa ___ 0

+ +

wU 0

m

0

(0

+

'I

cu 0 \100

00

w (U

<"0

0 F-

1-0 I-___ V_._ I_ -- U)

A AQ)

Iq

LL-I

T cu IUP 0

0

I LO

wl 0

N

CDCu* 0

w -0cc 0

M: 0

cc +

U T-4 - I uAL 0

APPENDIX 2

S PC-2000* AND

SPC-1 000

I DOCUMENTATION

IIII

S PC-20004IIIIIIUUIIBIU

APPLICATION REVISIONS

NIX ASSEMBLY USED ON LMr C- DESCRIPTION DATE APPROVED

320-557963 KU-BAND DS0 -- 00 INITIAL RELEASE 7-. .

AI

-.. -, - ! ,- -.SHEET 44 145 146 47 48 49 50 51 52 5 3 55 56157 58 59 61 62 63 64,REV-2212 }3 _

SHEET 23242526 230 32 33134 3536 37 38 39 40 41 42 43

REV _

SHEET.-,- 2 3\14 5 6 7 3 9 10 11 12 13 14 15 16 17 18 19 20 21 22"' W O TME w SPLIF110 AP14MALS. si, Tul We Day[

----- "A( -oVANTEK --.-TCNRAIr L.VELASOUEZ 121 /o

TOEACc* cmtcxto

u ,~b~faCl /,,P? fI*JCANICAL OPERATION SHEET

MATERIAL , CASE ASSEOABLY 5cP-ic/.oo

SIZE Ct ENT NO.

A 243MOS-59 /28I;-t As ig , I.

00 WT SCA IM •MXE l 9T O

MECHANICAL OPERATION SHEETPART #: 320-558373-001 REV W.O.# 5 3o/_73.0 g-PART NAME: CASE ASSY Stp& ooc

START 6 //7 /1 1 FINISH /STEP OPERATION DESCRIPTION IREFDOC REV OPER DATE CTY QTY

IN OUT

I. VERIFYB,. ; C;0-cl !, 4- It-il I o /0,I , - i . -

2. SOLDERRF FI" (2 PLACES) AP-0286CS J-fqz .. 7 I__o ! O_AO-557963 / /--7 /0

3. VAPOR DEGREASE PARTS PS-023886 ,2J92. /0 */O

4. FNE LEAK CHECK PS-029238 _0 110

S. VAPOR DEGREASE CASE PS-023886 l 1 0 1

6. 100% MFG INSPECT AWS-014355-800 17&5p 5 . /0 /0AD-557963

7. INSTALL RF/DC CONTACT (2 PLACES) 340-028489-001 /A//

AD-557963

S. EPOXY ThRUNE INTO CASE AD-57- ________ PS-505313 , .

9. INSTALLRFFILTERANDSMA AD-557963

10. GAP WEL CONTACTRIBBONTOTHRLINE PS501206 / //AD'-557963 / /0

11. CLEANCASE PS-023886 'c (o 1 /o

12. 100% MFG INSPECT AWS- /

014355- 800 '0'tq ((j (, LAD-557963- - -iq 1o1( o /

13. 100%OAINSPECT AD-557963JI-0235

S.ZE CODE MENT.0

0AVANTEK A_" 24539 Mos-55"/2 --SCALE WEIGHT SHEETOF REV

A.-

511 LU

cim,

0'0

Z zU

0 cc

.4 Il

00 0

It4f wl r4 C 4 0 r 1a -W7-n - 1 al

I-hi = i hi

mozo 0.

000

P- 4 19 u 94 04 A; )a ad a A. A.

'"" , 4 La U3 0 IF- '0 r0'4 0zh

HZ a .0 4 C) 0 V= O 0 44

I4 0 Ada PCO . U 4 0 a

01. to4 WW. 14 hp I-. a u U U w 0i

t. n U Iq bW 1 1 I l - 1 - % a WN

3 U *.a -o o 0 0 U 0 (0 -. 0

C' hi "a 4l hi U "- "' ma n ,i I N: C4 0 i 4 - .

r~0 n 4 e' - 0 a "4 "i hhI I In 0 a 0 N In 1 0 U U 0 .

ar 0 0 hi U 1 0 ( U

- -M

hi aT V. . I a - * CID 0 4 i

an 0 4a 0 0 0 0 0 0 0 o 01-. 01 0 00 0 0 0 0 0~ r 4 ( 4 0 C>-- 1 1 4 4

hilO~~ 0 - 0 i(0 IiI 0 4

g40 na 14 1 . 0 . ( v iu4( 0 i 0 0 i v 0v

93O 0 ' -i ai -

0

ri

all-.a- 4 O -. 1 . (0 9 0 0 4 U

00~

@00

z Z

092 J

CC -jIL Z) LLL

0 cc

W. 0

u 4 w W 94 94

00 0

p44

4 (7a-. z

. A.

hi ex~ 0S

.~A. 0 . I

000 0:;0~ 0U4 .

hi P, 0'

od 0 0. J .

i~0 =i 03 " I* 0 0 C.

pd i~d .40 Olb0,03 04 05 .I -l 0 A. ~ '

a~ 0m 0

PC0 M- 0

-0 04 I ow

IIU I

Ci

. 6 .+

* <- .. g " 4 4...

•~ S22 I++

1)00,

'I,,++ @ .

It:30

- + .__+ <+RL, -" - , -oL .3 .

-- .,.- .. O.ILI.. . .

.

r) CY

zI U

C..

IIJIN.

0

C? .0

I- U

.r, La

10

L, C:

U LJ

LiO

zL) W

0) a' L.J Z

C. p

Li A

0 0

0 Li

0 X

L, Q4 -,

-LJ M U I2 Z tn

z zI~ ~ Li r .(A C o <co-LJ C) C ,P-)0

AD-558373 I K//r/AD58373

C)

0C)

C)

II (4

3~AI .2 -4(in

Li ~

I- LO

0o V

10-

IUIII

SPC-1 000aIIIIIIIIIIIII

A..

100

CI.

LLJ 0

00 0) 00

I 4

00 000

0,0

I o 0..l4 -4. P-4 0 '

ca ,-4ca

log.- - . I b;~~0 to~

h4 JI( ui c M

00 00 w

0 1 " . ) .

Un 00. I -I p

00 4 1 0 0 0., o 0 Uv ,

P-0 0 I (2 m

0 b300 0 0

0g4 4 c

04 N

0 0

4t 4 0

AD-558128 I M/A/F/AD55OBIZR

U>-i cc C

(U-

-A 0V

ul I V I . I u V

C-)

(U

C,)I I-

aaLII-

Li

fuu

C~C

I m)

CL ma

I- Ia 0

z

R&D STATUS REPORT

DARPA Order No: 6268 Program Code No: 000008E20K43

)ntractor: Hughes Aircraft Co. as a subcontractor to SuperconductorTechnologies Inc.

°ntract No: N-00014-88-C-0713-S-HAC-1 Contract Amount: $737,991.00

Effective Date of Contract: 01 Sept 88

:incipal Investigator: Dennis Elwell

Telephone No: (714)759-7386

iort Title: The Processing of Hi-gh Temperature Ceramic SuperconductingDevices.

,porting Period: 10/1/88 - 9/31/91

FINAL REPORT

IMMARYThis report briefly summarises work at Hughes Aircraft as subcontractors toSuperconductor Technologies Inc. over the period of the contract. The aim of theivestigation was to evaluate two areas of possible application for high-Tc

Ljperconductors- connecting cables for IR detectors and passive microwavedevices- and to design, build and test a replacement device using STI'si terial. The most important contributions of this study have been the feedback( test data which helped STI in their outstanding improvements in materialperformance at microwave frequencies, and the development of an original methodfor tes ng microwave materials at cryogenic temperatures, which allows in-situkilibra-ion. The PIN-diode switched superconducting phase shifter which was

' lected for evaluation does not appear to be manufacturable at the presenttime.

TISCRIPTION OF PROGRESS

IR FOCAL PLANE INTERCONNECT

. detectors operate at cryogenic temperatures, and are connected to electroniccircuitry at substantially higher tempeiatures by a flexible cable fabricatedr. polyimide (see Fig. 1). The advantages of using a superconducting materialt carry signals from the detector to the signal processor are its low thermalconductivity combined with superior performance as an electrical conductor,including a relative absence of dispersion.

1I selected a typical cable in order to compare the predicted heat loss ofa superconducting equivalent with that of a cable using a conventionalr-tal. The device selected was from a program known as AOA and actually has1,ree cables and operates in a temperature range from 15 to 90K. In the caseoi the superconductor, deposition on polyimide is still not feasible and so weassumed a requirement for a rigid substrate. The total heat load for the three,nventional cables is around 60 mW for a temperature difference of 75K and able length of 5.5 inches. The dominant contributor to this heat load are the

two copper power leads, since the Kapton base is a poor thermal conductor. Ther-perconducting leads to carry the same current would impose a load of onlyi)out 0.3 mW and so the loading of a superconducting interconnect would be

SIUI

IIIIIII

Fig. 1 . Sketch of infra-red focal plane

array assembly.

IIIIU

aominated by that of the substrate. An ideal substrate material would be fusedsilica, which would result in about 8 mW of heat leak. There has been progressa the deposition of high-Tc films on silica with barrier layers and theatroduction of superconducting infra-red focal plane interconnects appears

feasible and could be considered for new programs, especially those where power-s of particular concern. Superconducting cables on substrates with higheriermal conductivity such as alumina would not be viable. There is a reluctance

Lor IRFP device engineers to replace flexible with rigid cable, so thepotential impact of superconducting cables would be particularly strong ifney could be flexible. Deposition on polyimide is difficult, although notupossible since temperatures for in-situ deposition are comparable with the

decomposition temperature of some polyimides. It is recommended to investigate'he development of superconducting films on flexible polyimide cables as theechnology for low-temperature (around 400 C) in-situ deposition matures.

zuch cables appear to have real potential in view of the extreme importance ofreducing heat loss in satellite-based IR detectors.

[CROWAVE MEASUREMENT TECHNIQUES -

'ing Resonatori the early phase of this investigation, the emphasis was o;. comparison of

Lne properties at microwave frequencies of STI's thallium superconductor withthose of conventional metals. A nominal goal for STI was the development ofaterials with X-band surface resistance an order of magnitude less than thatgold or silver at cryogenic temperatures. Sample size was limited to

0.5 x 0.5 inches, and the substrate was MgO.

ring resonator was selected for these weasurements on the basis of simplicitydad the low radiative loss. Various arrangements were tried for lainching thesignals into the sample and minimizing the influence of a strap between theimple and the launcher pad on the fixture. Our development of cryogenicicilities for microwave measurements on ring resonator samples was mainly

funded from our IR&D program and was described in a 1990 publication'D.J.Miehls, D.Elwell, P.S.Fleischner and A.A.Shapiro, Intl. J. Hybrid,icroelectronics 13#4(1990)85).

Several assumptions are necessary in extracting surface resistance data fromie microwave S parameter measurements. Among these are conversion from loaded) unloaded Q, subtraction of dielectric loss, and separation of microstrip

from ground plane contributions. A standard sample was fabricated from gold by9.rst sputtering a very thin Ti/W adhesion layer and then evaporating aboutmicron of gold. Gold was then plated to a total thickness of 9.5 microns,ell above the skin depth at 4.6 GHz. Loaded Q measurements on this standard

sample were 72.5 at room temperature and 193 at 33K. Our first TBCCO sample:om STI had a Q of 39.6 at 75K, 56.7 at 50K and 64.8 at 25K. These dataiggested a surface resistance for the superconductor of the same order as

that of gold at room temperature, much higher than the value expected from-xtrapolation of STI's high frequency cavity measurements. This disappointingrformance of the TBCCO film at lower microwave frequencies led STI to

,avestigate the origins of the effect and to modify their film depositionprocess to remedy the problem.

ie STI TBCCO film responded well to cycling between 20K and room temperature.There was no detectable degradation when a film was subjected to eight suche-vcles in an early test (Fig. 2). The variation in the resistivity near the:ansition temperatre is within the error of measurement.

TRL Measurements)llowing STI's breakthrough near the end of 1989 in producing greatly improved

'3CC0 films and the development of their own test facilities, our emphasis

0.1 E+ O0

0.1F.-O1

0.1E-02

E

Y, 0.1E-03

0

.1 1E-04 - 1st Temp Cycle

. - 2nd Temp Cycle.=.nu 0.IE-05

- - ---- 3rd Temp Cycle

- - - 4th Temp Cycle

0.1E-06 - 5th Temp Cycle

- ,6th Temp Cycle

0.1E-07 -. 7th Temp Cycle

.. --- 8th Temp Cycle

0.1E-08 I I"

70 80 4 -' 0 100 110 120 130 140I5Temperature (kelvin)

5 Figure2,. Resistivity vs. temperature data for TBCCO subjected

to 8 cycles from 293 - 20K.

IIIIII

I

switched from characterizing materials towards the design and fabrication of aicrowave device using the STI superconducting films. We selected a PIN-diodeditched phase shifter which is the key component in airborne phased array

radars. Insertion loss is of fundamental importance in respect both of the-adar range and the power dissipated.

wecause of the very small insertion loss which can be achieved even inconventional phase shifters, the introduction of replacement superconductingavices requires very sensitive and reliable measurement techniques for7yogenic temperatures at microwave frequencies. Our initial experience

of X-band measurements on high-Tc materials convinced us that measurements'hich involved comparison between conventional and superconducting samplesKchanged at room temperature would not be acceptable; allowing the chamber.o warm to room temperature and then re-cooling would introduce uncertaintiesof the same order as the differences to be measured. A method of in-situilibration of the measurement system, with immediate test of theiperconductor at the cryogenic temperature, was therefore devised and a

cryochamber modified to allow the introduction of this technique.

i important aspect of our measurement technique is the absence of direct,ntact between the microwave input and output signal carriers and thesuperconducting sample. RF coupling is achieved by a dipole/balun arrangement,ith the RF energy entering the cryochamber via a ridged waveguide (See Fig. 3).ilibration is achieved by a TRL (thru reflect line) arrangement, the various

elements in this array being moved into position at the measurement temperaturehv the use of a slider, as shown in Fig 3. Although the ridged guide inirticular constitutes a significant heat sink, it was possible after

Lareful modifications to reach temperatures as low as 29K. Precise location ofthe sample and standards in both horizontal and vertical directions is difficult) achieve to the required accuracy because of thermal expansion and)ntraction of the metal, but this was an important segment of our later work.

-ha system was validated using a thick film silver test part incorporating a:andard mismatch circuit fabricated on an alumina substrate using thick filmlver. This consisted of a microstrip impedance transformer with a return loss

of about 10db. Fig. 4 shows scattering parameter data taken at room!mperature. The data is all taken in a frequency range from 9.5 to 10.0Iz. Fig. 4a shows insertion loss and return loss; both parameters show a small

dependence on frequency. The markers are at values of 0.34 db for insertionloss and 14.35 db for return loss. In Fia. 4b we compare loss data for the Thru,id Delay lines. The Thru line has lower loss, especially at the lower.:equencies, but the return losses are well macched at reasonable values.Fig. 4c shows in the lower trace the return loss of the short circuit. Theilue at the marker is about 0.01 db which is considered acceptable. Also shown; the short to short isolation. we considered a value above 40 db to be

acceptable and the measured values were around 60 db. There is a trend for theisolation to decrease slowly at the higher frequencies, the marker being at33.9 db. In order to obtain the data of Fig. 4, we had to solve serious

_.:yochamber moding problems. The use of absorbent materials in appropriatelocations reduced this problem to a reasonable level, without seriously:fecting vacuum or minimum achievable temperature.

Fig. 5 shows the data for the same standard mismatch circuit measured at 60K.Fig. 5a shows the insertion loss of the mismatch line and the data are!nerally similar to those at room temperature, with rather lower insertion

_iss at the higher frequencies. Fig. 5b compares the Thru and Delay lines andreveals a new problem, namely microphonics due to vibration of the compressor)tor. This vibration has a strong influence on the return loss and has greatly)ntributed to the noise level in this figure. In our CTI cryochamber, the cold

head is located directly above the compressor and it is impractical to insert

T) IERIAAL ISOLATIONCOUPLER FORR~OTARlY O1IVC!

VACUUMro /EL

TOTOPOEWA - STANOARDS

SEIO ,3Aprt o irwaeSprmee euurmnsacrywonictmeaue sn RLclbainsadrs

n fIo~ v MAC, o~ Y Aini:F 0. 0 do nJ:, 6.6 do

0 dO/ 5.6a dO/--0.3444 do -"14.345 dO" ...... . ... .- - .. 1. . -

...... --L~ : ... -- T--

S a)

START 0.566e00600 OHZ 571 M1 go MAO 51 & 1 Z gWAO

STOP 16.66000606 OH HLP-' ou dU 141111 -0. aiu0.1 dO/ A 26.0 dII'

-0.2236 dO 1 -47.005 gG

-, I I

_ I

521 log MAC 511 log MAO START 0.500006060 (j41

REP 6.0 in REP 0. do STOP 10.ogooogggo O<E20.0 go/ 6.1 do/

-53.0-d -6.0.35 d.

A~ Fig. 14.Validation test fixturedata: (a)Insertion loss and

.. .... return loss for lUdb mismatch.

I i I kb)jnsertion and return lossIfor Thru () and Delay (D) line.kc)-nsertion loss ano isolation,L , -- j" + '+ .-' --- . - for Short.

_ - I I .. .. . . . - I. .

L . .. 1 .. .. . .. .. . . ... .. . ... . .. ; .

TAPlT VJ,500000000 U14°TeV' IZ.O0000OO00 DO

I:II 0. 0 dO fliPr 0. a dO

16 1 do/ 5.0 do/.. 3551 do -.12.050 dO

AA

Fift~h I-a

STg AT .50000 O yM TRT 356000~~zi~~RE 0. do 89P 0. 0 d T oeee o

20.0 do/ 0.1 do/

Fig. q-Data for same sample

.... :,- +

... . in Fig. 6 .. at 60K.

5TISRT J,5CU0O00OU Ui; -5TCY. 1OOOO0a 0141

vibration-damping material between the cold source and the sample. In caseswhere the duration of the measurement is short, the compressor can be switchedff during measurement but the temperature then rises fairly rapidly. The Sarameter measurements normally take too long (about 10 minutes) for this to be

a practical solution, but we were able to reduce the noise by introducingdditional smoothing of the data. The difference in insertion loss betweenhru and Delay lines was unreasonably high in this test. This difference was

probably due to a positional error as described above. An additional source oferror is that the dipoles may not be exactly similar in size and shape due tohe manufacturing tolerances of thick film technology and this could lead toifferences like that shown in Fig. 5b. This problem can be reduced by verycareful processing and is probably not significant in this case since it doesot show up in the room temperature comparison, but it must be considered innterpreting the data. Fig. 5c shows data for the Short, which is excellent

ror both insertion loss and isolation; both are very flat across the frequencyrange.

ae reproducibility of the measurement system at 60K was checked by a series ofsix runs with the results plotted on the same graph. Fig. 6 shows this data.-h each case the measurement was taken through the complete calibrationocedure then the loss of the Delay line was measured and plotted. The sample

was the same as discussed above. The reproducibility was generally veryaood except for the first run which showed rather higher loss at the lower:equencies.

Phase Bit.nce the main goal of this project was the comparison between normal metal andiperconducting phase bits, the emphasis after the initial tests was to make

and test a phase bit. Thin film gold was chosen for the standard sample. A90 degree phase bit was designed on a substrate with dielectric constant K-25;imulating lanthanum aluminate) and fabricated on a 1 cm square ceramic, shown in Fig. 7a. This phase bit is dropped into a hole in the TRL substrate(see Fig. 7b) for measurement, after connection to the conductor strips byY idging straps of minimal height.

Tte substrate fabrication technique is a multistep process using conventionalthin film technology. The substrate surface is first coated with a thin layerc titanium/tungsten to give good adhesion. This is sputter-coated with goldE.d then electroplated to a thickness of about 3 microns, substantially thickerthan the skin depth at X band. A 4210 photoresist is then used to pattern ther-tal coating by baking, exposure, development and etching, followed byc eaning to remove residual pnotoresist. A positive photoresist could be usedas an alternative procedure. Connections to the back side of the substrateare preferably made by via connections through holes in the substrate.L trasonic drilling is a satisfactory method of putting via holes in LaAIO3s bstrates but the presence of the via holes complicates the film depositionprocess in the case where the sample is superconductor rather than gold.

r' asurements on the standard phase bit at room temperature are shown in Fig. 8.Fig. 8a shows that the phase shift is around 94.5 degreps and has only a slightvariation across the frequency t3nd. This shift could be corrected and thef equency dependence removed Jy design iteration, but these performance figuresa e considered satisfactory for this initial demonstration. Figs. 8b and 8c aresimilar except that the signal is applied at opposite ends of the chamber, sothat the data are a test of the reversibility of the arrangement. Both chartss ow similar behavior, .ith a tendendy for the insertion losses to diverge withiicreasing frequency. This is typical behavior for this type of phase bittopology.

F 3. 9 shows data for the phase bit at 30K. The phase shift is rather higher

S2 1 I oq MAGREF 0. 0 dB

0. 1 dB/

hp

A

_ _ _ _ _ _ __

_ _ _ _ _ _ _

START 9.50000e00 GHzSTOP' 10. 00000CZ00 C3Hz

Fig. 6. Repeatabi li ty test data at 6uK. insertion loss of ludbmismatch repeatea 6 times at tbOY.

o Ca)

RE .

1I F'*H *

IWIi-f-

START 9.59"0e0o GSO l.aNHSO O0 Ga S1 & S14 MI gHA S - QL C og

-o.g s U F 6. -,8:.S dS is. d-1.462 ds -25.547 d8

(b) D1

R!U

EV..E .

A- -

snit; o~l.....i andt return

£141,NL I~ 2 IQ I.g - -- "

-START

I 1.63 ~ST 1STOPeee i

I--------------------- Fig.8 . Data for 9u degree phase Dit9.:'- - " - ... .. at room, temperature: Ca) IPhase

S -- - snift; (o)Insertion and return----------------------- loss; (cjSimilar to (b) but withI-----------------------------connections reversed.

= - - -- -l-

TARq 1.. O0OOO0d0 r",,

S 2 1 /MI LREF 0. 0 01 45.0 0/

V 104.71 _

hp

*

ASI

- -, -

! *1

1......y.. ......- -re

START 9.508600 GHzSTOP 19.9606980 GHz

S 2 1 & M1 Iog MAG S 1 1 & M2 log MAGREF 0.0 dB REF 9.8 dB1 8.5 dB/ A 10.0 dB/V -1.9888 dB 1 -22.894 dB

*

AS

D

--

START 9.500000000 GHzSTOP 10.000003000 GHz

Fig. -. (a) Phase Shift; (b) Insertion loss(upper data) and return loss (lowerdata) for standard gold phase bit at 30K.

than that at room temperdture (104.7 degrees at center frequency) and thei sertion loss is a little higher than at room temperature (1.59db versus1 51db at room temperature). This appears an anomalous result since theconductivity of gold must increase at cryogenic temperatures. However ther-turn loss of the cooled bit is less than the return loss of the roomt mperature bit so this would contribute to the loss. Some variablity wase~perienced which was attributed to physical mismatch between the printeddipoles on the substrate and on the ridged waveguide. This problem could ber iuced by the introduction of a sensitive position locator to check on smalld splacements on cooling so corrections could be made before measurements weretaken. Unfortunately the program was terminated before this modification wasc-mplete.

SURVEYS OF HIGH-Tc APPLICATIONS

0 er the course of the contract, there were three initiatives within HughesA rcraft Company to seek promising applications for high-Tc materials inmicrowave systems, particulaily radars. The microwave field was consideredb- Hughes to be the potentially the second most important applications area fort e new superconductors, following data processing of IRFP array data usingJusephson technology if this can be developed in these materials.

T ? first initiative was our survey which was reported in the Status Report of2 28/89. A telephone survey of Radar Systems Group (RSG), Ground Systems Group(GSG) and Space and Communications Group (SCG) engineers found at that time theg-satest interest from the latter group. Calculations had been performed at SCGo a narrow-band downconverter operating over a wide frequency range from a fewMnz to 1.5 GHz, and substantial performance improvements coupled with weightreduction had been calculated. Unfortunately the device was for a classifieda olication. We held a follow-up meeting at RSG and were encouraged to pursues Derconducting phase shifters since reduced insertion loss would allow greaterradar range and so provide an important tactical advantage which could moret1'an compensate for the cryogenic cooling requirements.

A. superconducting technology matured and material improvements approachedeven the more optimistic predictions of microwave performance, our systemsc [leagues took a harder look at their earlier projections and decided thata zooling system to maintain a phased array at 77K in a modern fighter wasimpractical in view of already severe power and weight restrictions.

H ;hes carried out in May 1990 an applications study aimed primarily at ground-based applications of superconductors where power and weight considerations arenot so severe. High sensitivity receivers were seen as the most likelya Aication area, especially at the lower frequencies in view of the quadraticd )endence of surface resistance on frequency. No single strong candidate forshort-term HTS application was identified in this internal study.

I March 1991 a presentation of STI technology at RSG was arranged, and wasa.cended by several senior engineers and managers. A low noise exciter wasidentified in this meeting as the leading candidate for replacement of ac iventional by a superconducting device. This has the advantage over thep ised array that only a single circuit is required per system, so greatlyreducing cryogenic requirements. However this proposed application wasiaentified too late to substitute for the phase shifter in this project.

C%.4CLUSIONS

T s project was terminated by STI prematurely before the improved performanceo a TBCCO superconducting phase bit could be quantified. Prior to thisdecision, progress both on our measurement facility and on superconducting

test samples was slow in comparison with our expectation. We underestimated the-*iysical problems of implementing cryogenic measurements using the TRL method,i d the presence of via holes in the substrates presented new challenges toSTI's film deposition process that led to delays in sample preparation. Inretrospect, an arrangement in which we relied on STI to provide samples ini rect competition with their own internal requirements, at a time when theL position process was in a phase of intense development and samples werescarce, was not a formula for the most rapid progress.

1 recommend that the cryogenic TRL measurement system be optimized and willattempt to do this with internal Hughes funds.

1 must reluctantly accept that substitution of HTS devices into majorLstems will be slower than we had expected, since the most important thrustwithin the systems houses is currently towards cost reduction. The cost of,yogenics in systems which have not previously required low temperaturet)eration is a major .hurdle, in- spite of a slow expansion in cryoelectronicsgenerally. A superconducting phased array radar now seems a very distant goal,but presumably one which will be realised at some point in the future. In the:iorter term, we recommend that microwave HTS development be focussed on smallt mponents which can give substantial performance improvements with only localcooling, such as the low noise exciter referred to above.

FINAL REPORT

CONTRACT NO. N00019-88-C-0173-S-SA-1

15 NOVEMBER 1991

Laockheed Sanders, Inc

U PROGRAM SUMMARY

U Superconducting Technologies, Inc. and Lockheed Sanders are developing a passive microwave

device incorporating high temperature superconducting (HTS) materials. The program includes

development, characterization, lithography, and patterning of superconductor materials. An

inverted microstrip HTS resonator is used to characterize a microwave oscillator.

U ACCOMPLISHMENTS

* The accomplishments over the life of the contract can be grouped into three areas: HTS

characterization, oscillator development, and HTS lithography. The most significant advances3 were made in the processing techniques used to pattern HTS materials. The HTS

characterization work was used to support the lithography effort in the early stages of the

3 contract. The characterization was used to determine if a particular lithographic process was

harmful to the HTS material. A 10 GHz oscillator using a half wavelength resonator was

3 designed and tested with a gold resonator at both 298 K and 77 K. This work was transferred

to STI before a working HTS resonator could be inserted into the oscillator.UHTS CHARACTERIZATIONUIn the early stages of the contract, the HTS films were characterized with a surface profilometer

3 and a 10 GHz parallel plate cavity. The surface profilometer measurements gave an indication

of the thickness uniformity of an HTS film. Early in the contract the surface of the HTS material

3 varied by as much as ±1 [tm. As the film quality improved, surface profilometry measurements

were performed less frequently. The parallel plate cavity consisted of a copper ground plane, a

3 2 mil thick polyethylene dielectric, and a coaxial input probe. The HTS sample was placed faced

down on the dielectric next to the coaxial input to form the cavity. The parallel plate cavity is

shown in Figure 1. The cavity measurements were a easy, quick way to gauge the effect of

process steps on the HTS material. A gold standard measured in the cavity produced a Q of -80

while the best HTS films would produce a Q of -400.

II

II HTS OSCILLATOR

S A 10 GHz oscillator was designed using a half wavelength resonator as a stabilizing element

which is shown in Figure 2. By using a high Q resonator, the performance of the oscillator

3 would be improved. A resonator mask was designed and both gold and HTS resonators were

fabricated. Because of the poor HTS film quality at the beginning of the contract the gold and

HTS resonators had about the same Q. An oscillator using the gold resonator was tested at 298

K and 77 K and had performance as expected. Eventually STI supplied an HTS resonator and

ground plane with a Q of 4000 to be used in the oscillator. These were installed but the

oscillator did not fire off. Subsequently it was found the Q of the HTS resonator had fallen 150.

The cause of the degradation was never determined and all the hardware was transferred back

to STI to complete this task.

LITHOGRAPHY

Over the life of the contract many lithographic, passivation, and normal metal contact techniques

were evaluated. In addition air bridges were developed and substrate thinning experiments were

performed. After a series of materials evaluations, the wet etch technique chosen was a 1:200

HCl/I- 20 solution at room temperature. Shipley 1375 was used as the photoresist with MIF312

as the developer. Exposure was accomplished with 360 nm UV mask aligner. Two methods for

bondable contacts were developed. The first was used for bonding on top of HTS material. A

5% bromine in methanol by volume solution was used to etch the surface of the HTS material.

Sputtering 1 [im of gold on the prepared surface will produce bond pull strengths exceeding 20

g. The second method was used for bonding to an isolated pad on LaA10 3 which was then

connected to the HTS material with a gold lift-off interconnect. The isolated pad consisted of

an adhesion pad of sputtered TiW/Au followed by a 1 [tm evaporated Au layer. As an alternative

to wet processing, ion beam milling experiments were also performed. A Commonwealth

Scientific 3" Ion Mill System was used to pattern some test circuits. Typical parameters were

500 V accelerator voltage, 500 mA/cm2 ion beam current, and 2"10 4 Torr gas pressure using

argon as the ion species. This produced circuits with improved edge definition but no discemable

improvement in rf performance over wet etch techniques. The contrast between ion milling and

Iwet etching can be seen in Figure 3. Contacts electroplated directly to HTS material were was

also evaluated, but were discarded because the electroplating process dissolves the HTS material.3 Air bridge technology was also used and transferred to STI. Since part of the air bridge is

electroplated, a 5 level mask scheme was developed to isolate the plated portion of the air bridge3 from the HTS material. The freestanding bridge is then connected to the HTS material. An

example of a gold air bridge over HTS is shown in Figure 4. Finally thinning of LaA103

substrates was performed. One square cm samples were successfully thinned from 20 mil to 5

and 4 mil thicknesses. A sample was thinned to 2 mil but the grinding induced a severe camber

in the substrate.

III

IUUIII

I U

*I-

I0

* -

*IIUIIII

IIdUc

1~ -r o

I' .. ,reaa. 1116.

IIIIIIIIIII *

I b

I* -.

II U

IFigure 3. Wet etched thallium 1-ITS (top), ion milled YBCO (bottom).

I

IT

IcIc

7f

IA

ILMSC P008929

FINAL REPORT

* SUPERCONDUCTOR APPLICATION RESEARCH

October 15, 1991

U R. J. Adler and W. W. Anderson

II

Prepared for

SUPERCONDUCTOR TECHNOLOGIES, INC.SANTA BARBARA, CA

Contract No. N00014-88-C-0713-S-LMSC-1

ILOCKHEED RESEARCH & DEVELOPMENT DIVISION

LOCKHEED MISSILES & SPACE COMPANY, INC.PALO ALTO, CALIFORNIA 94304

IIIIIII

I

LMSC P008929

1 1. INTRODUCTION AND SUMMARY (PART I)

3 Heat leaking along the wiring into a low temperature satellite system, e.g.an infrared sensor at 10 K, can cause rapid loss of coolant or highrefrigerator power consumption. This can shorten orbital lifetime andcomplicate the payload. If a set of high temperature superconductor(HTSC) leads is inserted between the cold system and the rest of thesatellite, the heat leakage can be greatly reduced, while no extraelectrical resistance is added. A superconducting thermal isolator device(STID) consists of a substrate of low thermal conductivity on which HTSCtraces are deposited. A low thermal conductivity housing holds theassembly and provides electrical connections.

Five STID prototypes have been built and tested at LMSC using substratesand traces from STI. They use TIBaCaCuO traces on a polycrystalline3 zirconia substrate. Their thermal isolating performance is excellent, buttheir current carrying capacity is somewhat too small for practical3 applications: the traces carry a few hundreds of amps per cm 2.

Five Mark I electrical test devices have also been built and tested. Theseuse TlBaCaCuO traces on a LaAIO 3 substrate, which is not a good thermalinsulating material; they are intended only for electrical testing. Theyhave excellent electrical properties, with traces carrying hundreds ofthousands of amps per cm 2 .

I Finally two Mark II electrical test devices have been built and tested.These use TIBaCaCuO traces on substrates of single crystal ZrO 2 and

I polycrystalline ZrO 2. The single crystal Mark 11 device is quite superior to

the polycrystalline device, and has adequate electrical properties, withtraces carrying about 5,000 amps per cm 2 . The single crystal Mark II test

device was monitored over a period of about 5 weeks, and the tracequality deteriorated significantly over this time. This indicates that apassivation coating is probably necessary.

I

I

I

LMSC P008929

2. REQUIREMENTS BACKGROUND

2.1 The Problem

Figure 1 show, a schematic of a typical satellite low temperature system.

~ K 8AD IA T IO N0 K

rCONDUC1lO

FOCAL PLANE IR SYSi Eivi: EXTERNAL SPACECRAFT

SURVEILLANCE OR ASTRONOMY SYSTEMS AND SHROUD

Figure 1. Schematic sketch of typical satellite low temperature system.

Examples might be an infrared surveillance sensor system, an infraredastronomical telescope, or a research spectrometer. The cold parts ofsuch systems typically operate at under 10 K. They must communicatewith the rest of the spacecraft, typically at about 77 K, through leadswhich pass power, command signals, and output signals. The leadsinevitably leak some heat by conduction, and also generate some heatthemselves due to their internal resistance, referred to as Joule heating.As a result there can be a serious loss of coolant in a stored cryogensystem or power consumption in a mechanical cooler. This generally

2

LMSC P008929

translates into shortened orbital lifetime for a given weight in orbit, anda larger and a more complex payload. In turn this implies that megadollarsmay be wasted in space.

We might think to solve the problem by making a clever choice of metalfor the wire leads, to pass electrical current but block heat flow. Thiscannot be done however, since normal metals which are good electricalconductors are also good heat conductors. This fact is embodied in theWeidemann-Franz relation, equation [1], which gives the ratio of thermalto electrical conductivity as a function of only the temperature T and twouniversal constants, the electron charge e, and Boltzmann's constant k.

Kth /KceL = (Trk/e)2 T/3 (Weidemann-Franz relation) [1]

The Weidemann-Franz relation should hold reasonably well for anymaterial in which electrical current and thermal energy are transportedby normal electrons. It is reasonably accurate for most metals; for

example around 00 C it holds to about 25%. (ref. 1)

2.2 The Conceptual Solution

Superconductors are not normal conductors and do not obe-y theWeidemann-Franz relation; electrical conduction is by condensed Cooperpairs, which behave like a condensed gas of bosons and do not transportentropy or heat (ref.s 2, 3). It is particularly fortunate that hightemperature superconductors (HTSCs) can operate at the upper end of thetemperature region desired, about 77 K, the boiling point of liquidnitrogen. By inserting a HTSC between the hot and cold parts of the systemthe heat leakage problem can be ameliorated. Unfortunately HTSCs cannotyet be made into usable flexible wires, so we need a device such as shownin Figure 2; this has HTSC traces on a dielectric substrate, which must bechosen to be a good match for the HTSC crystal structure, and also a goodthermal insulating material. In addition the substrate and traces must bemounted in a low thermal conductivity housing.

3

II

LMSC P008929I

WIRE BUNDLE WIRE BUNDLE

I R ______EXTERNAL

SYSTEM SYSTEM

I HTSC TRACES ON SUBSTRATE

IFigure 2. STID concept.

I The detailed design for a STID will be discussed in section 6.

3II3III3£3 4

I

LMSC P008929

3. THERMAL CONDUCTIVITY OF SOME MATERIALS

The substrate material should be strong and have low thermal

conductivity. Figure 3 compares the thermal conductivity of copper, ZrO 2

(zirconia), and the popular substrate material LaAIO 3. At liquid nitrogen

temperature, 77 K, the conductivity of ZrO 2 is 20 times less than that of

LaAIO 3 and 300 times less than that of copper. At the low temperature of

10 K it is respectively 100 times and 30,000 times less.

1000000

j1000__I__ I:Thermal A - - Cu

Conduct. 1000* M * LaAIO3(mW/cm K)a

I "1 * a~ ZrO2:Y100-

1 0 fAt I___Soo Ilf I,,,, .__z -

1 0 100 1000

Temperature (K)

Figure 3. Thermal conductivity of copper, LaAIO3 , and ceramic Zr0 2 .

We may conclude that Zr0 2 offers very low thermal conductivity. It is also

a particularly hard and strong material and therefore offers an attractive

substrate choice.

Thp data in Figure 3 are for polycrystalline or ceramic ZrO 2, which was

used in our prototype devices. We expect the thermal conductivity of

single crystal ZrO 2 to be comparable to the polycrystalline material. This

5

II

LMSC P008929

should be true in particular at low temperatures where long wavelengthphonons are most important, and grain boundaries offer less thermal

3impedance.The conductivities of various HTSCs have been measured to be in therange 10 - 100 mW/cm K, comparable to ZrO 2. The traces in a STID havea very small cross sectional area compared to the substrate sincethey are only about 1 pm thick. We may conclude that the heat leakagethrough the superconducting traces should be negligible.

I6

II

LMSC P008929

£ 4. HEAT LEAKAGE EXAMPLES

I4.1 General

Each cryogenic space system has its own individual properties, so it is notreasonable to attempt a general analysis of heat leakage. We will insteadconsider some illustrative examples of the leakage of heat into a coldsystem via metallic wires compared to a STID structure. As noted abovethe HTSC leads should have negligible leakage compared to the substrate;we will assume that the housing for the STID can also be made withleakage that is negligible compared to the substrate

I There are three sources of heat leakage into a cold system: (1) conductionfrom the hot exterior world which we take to be at about 77 K, (2) JouleIheating due to current flowing in the metallic wires, and (3) radiationdirectly from the exterior world. Radiation will in general be rather smallfor a hot exterior at only 77 K, which is hot only by comparison with the

I cold system at about 10 K; the reason for this is that the radiation rate isproportional to T 4 (ref. 4). In any case the radiation leakage is common tothe use of wires or a STID. We will thus consider only conduction andJoule heating in the following discussion.

The largest leads into a cold system generally provide power, andtypically may carry about 500 mA. Other leads carry command and sensorsignals into and out of the cold system, both at generally small currents,say of order 20 mA. For this reason the power leads are the mostimportant source of Joule heating. A portion of the power carried by theleads will also be dissipated in the cold system itself, and the use of aSTID clearly cannot prevent this.

3The size of the signal leads is often not determined by the current theymust carry but by mechanical strength considerations; that is they are the

3 so-called minimum gauge, which is usually of order of 0.003 cm diameter.

Finally it is important to realize that all of the leads may have a short

£7

LMSC P008929

duty cycle; that is they may carry current for only a small fraction of thetime, but the leads must remain connected at all times. For these reasonsthe main source of heat leakage is generally conduction, and this is whatwe will illustrate below.

A further word about Joule heating is in order. A very thick wire will leakmuch heat by conduction, but will not produce much Joule heating becauseof its low resistance. A very thin wire will not leak much heat but willproduce much Joule heating because of its high resistance. At someintermediate wire size the heat leakage should be a minimum, with nottoo much heating by either conduction or Joule heating. The determinationof the optimum wire size is a straight-forward problem but depends onthe temperature range and the wire materials, and usually must be donenumerically.

4.2 High Current Leads

Let us now consider some examples of how one estimates heat leakage byconduction and see why a STID should be expected to reduce it. First wewill consider an example of high current carrying leads, a bundle of 10copper wires able to carry 500 mA each, and about 5 cm long. We assumethe cold system is at 6 K and the hot environment is at 77.5 K, whichcorrespond to a laboratory test setup that we have used. Figure 4 shows aschematic of the wire bundle and the relevant numbers.

The only physics equation that we need is that giving heat flow in units ofpower as a function of the temperature difference (ref. 5).

P = Kth (A/L) AT [2]

Here the power P has units of watts, A is the cross section of the wire, Lis the length of the wire, AT is the temperature difference, and Kth is thethermal conductivity. If the conductivity varies with temperature, as itusually does, the appropriate value to use is the average over thetemperature range. (This is shown in Appendix A.)

8

I

LMSC P008929

5 cm

5.0 X10 5 cm2 10 A/cm 2

(a) 10 Copper Wires @ 500 mA each

.015 cm 1.59 cm

E2 2U LO3.81X102 cm2•

* to

(b) STID Substrate and Traces

62 5 cm5.1 X10 6 cm 2 u 5c

(c) 20 Copper Wires, 1 mil diameter

Figure 4. Schematic for heat leakage comparisons.

To determine the area of the wire needed for this example we use a fairly

realistic rule-of-thumb, that the the current density in the wire should

9

I

LMSC P008929

I not exceed about 5,000 to 10,000 A/cm 2 . (We cannot apply a wire sizeoptimization procedure since the duty cycle is not given.) By this currentdensity criterion the area of each of the wires must be 5 X 10-5 cm 2 . Theaverage thermal conductivity of copper can be estimated from the curve inFigure 3 to be about 47 X 103 mW/cm K.

The heat leakage through 1 wire and the bundle of 10 are:

I P(1) = 33.6 mW and P(10) = 336 mW [3]

This rate of heat leakage implies that for each day in orbit about 29 x 10 3

J or 6.9 x 103 cal leak into the cold system, which is a sizeable amount.

4.3 STID

3 A STiD could be constructed and inserted into the system of leads asindicated in Figure 4. The dimensions of the substrate shown in Figure 4are those of our prototype design: 1.59 cm long by 2.54 cm wide by .015cm thick. The average thermal conductivity of ZrO 2 is about 15 mW/cm K,as can be read from the graph in Figure 3. Ignoring the very small leakagethrough the traces themselves we obtain from equation [2].

P(S) = 25.7 mW [4]

The reduction from the 336 mW for the copper wire bundle is over an order

3of magnitude

34.4 Low Current Leads

Next we will consider an example with low current leads, a bundle of 205 copper wires to carry 20 mA input or output signals, about 5 cm in length.The size of such wires is not usually determined by considerations of3 current density but by mechanical strength. We will assume that 1 mil

S10

I

LMSC P008929

copper wires are used, with a radius of 1.27x1 0 - 3 cm and an area of5.1x10 - 6 cm 2. Again using equation [2] we obtain leakages through 1 wire

and the bundle of 20.

P(1) = 3.4 mW and P(20) = 68 mW [5]

If a STID were inserted into the wire bundle the leakage would be theI same as previously obtained in equation [4], or about 25.7 mW. Thus there

is a reduction of leakage to be gained, but it is much less significant thanfor the heavier current leads first considered.

Table 1 gives a summary of the results of this section.

Table 1: Summary of Heat Leakage Examples

3 Specific Thermal Area Length PowerCase Conductivity Leakage

(mW/cm K) (cm 2) (cm) (mW)

High Current 47 X 103 5.0 X 10- 4 5.0 336

10 Cu wires

31 Low Current 47 X 1 5.1 X 106 5.0 68

20 Cu wires

STID 15 3.81 X 10-2 1.59 26

IIII

I'

LMSC P008929

5. PAYOFFS

5.1 Orbital Lifetime Increase

To estimate heat leakages in real satellite systems one makescalculations like those illustrated above, but including Joule heating,variation of thermal conductivity with temperature, and a considerationof wire size optimization. Together with researchers in cryogenicspacecraft systems at LMSC under the leadership of T. Nast we have madeestimates of heat leakage for 3 cryogenic spacecraft, and have therebyestimated the benefits in terms of orbital lifetime increase. We discussthese in the following paragraphs and summarize the results in table 2.

The spacecraft and systems studied were:

(1) Cryogenic Limb Array Etalon Spectrometer(2) Advanced ".y Astronomical Facility Spectrometer(3) Spirit III - iler

We will dis .Lss each in turn.

(i) Cryogenic Limb Array Spectrometer (CLAES):

CLAES is a satellite-borne instrument designed to study the atmosphereof the earth by detecting and measuring its gas composition. It is beingflown on the Upper Atmosphere Research Satellite (UARS), launched in fall1991. The CLAES device, including instrument and cryogenics, is about 2 mhigh by 1 m wide. The mass of the instrument is about 300 kg, and themass of the cryogenic cooling system is about 900 kg. Note that the massof the cooling system greatly exceeds the mass of the instrument, whichis typical of cryogenic satellites. An array of semiconductor chips is usedas a sensor, and is maintained at 15 K. Sensor cooling is by solid Ne at 13K, while solid CO 2 at 125 K cools the optics and some peripheral

equipment. Temperature of the instrument shell is about 200 K, andspacecraft temperature is about 300 K, which is typical for most

12

I

I LMSC P008929

3 satellites. The present system has a heat load of 783 mW into the sensorpackage, and an expected useful lifetime in orbit of 1.8 years.

If CLAES were fitted with a STID we estimate that the heat load would bereduced to 614 mW and the useful lifetime in orbit would increase to 2.3years. The benefit is thus about .5 year or 28% extended lifetime. (SeeBenefits Table.)

(2) Advanced X-Ray Astronomical Facility (AXAF) X-Ray Spectrometer:

The AXAF is designed to do basic X-Ray astronomy from orbit. It is in thepreliminary proposal and contracting stage, and should fly in about 5 to 7years. The X-ray spectrometer is the primary instrument on the satellite.With cryogenic system included the size of the instrument package isabout 2 m by 1.7 m, and the mass is about 350 kg; the mass of the actualinstrument is negligible compared to the cooling system.

Detectors are Si chips at .1 K. These are cooled with superfluid He to 1.5K, then to .1 K by adiabatic demagnetization. A mechanical refrigeratorcools the outer shell. The heat load of the spectrometer is about 6.2 mW,and the lifetime is about 6.1 years.

If the spectrometer were fitted with a STID we estimate that the heatload would decrease to 5.7 mW and the lifetime would increase to 6.6years. The benefit is thus about .5 year, or 8%. (See Benefits Table.)

(3) Spirit II:

Spirit III involves an infrared telescope and a large cryogenic coolingsystem. The satellite will do a combination of surveillance and science3l missions. It is in the fabrication and assembly stage, and launch should bein about 1993. The detector is a semiconductor focal plane arraymaintained at 10 K. It is cooled by a 1000 liter tank of solid H2 at about

20 K. The instrument and cooling system shell are maintained at about 200K. Size and mass of the system are similar to the CLAES system discussed

I1| I

I

ILMSC P008929

3 above. Heat load of the instrument is about 694 mW, and lifetime is about1.8 years.

I If Spirit III were fitted with a STID we estimate that the heat load woulddecrease to 612 mW and the lifetime would increase to about 2.0 years.5 The benefit is thus about .2 years or 11%. (See Benefits Table.)

5Table 2: Benefits Table

Program Present Design With STID Benefit5 Name Heat Load Life Heat Load Life

Cryogenic Limb 783 mW 1.8 yr 614 mW 2.3 yr .5 yr (28%)Array EtalonSpectrometer

AXAF X Ray 6.2 mW 6.1 yr 5.7 mW 6.6 yr .5 yr (8%)* Spectrometer

Spirit III 694 mW 1.8 yr 612 mW 2.0 yr .2 yr (11%)

*5.2 Benefits Table

The table summarizes the benefits of eliminating the heat load due to the3presence of electrical lead wires for three projects that are being studied

or developed at LMSC. The heat loads and lifetime numbers are based ondetailed system analysis and include the benefits of thermal groundingand vapor cooling effects; that is full benefit has been taken of vaporcooling leads. Joule heating in the leads, thermal conduction through the3 leads, and the interaction of these effects has been included in thecalculations.

13

| 14

I

I

LMSC P008929

55.3 Qualitative Design improvement

3 Some satellite systems would benefit from the use of a STID by having aqualitatively improved design, completely unlike present concepts. Theastromag is such a system.

The astromag project involves a pair of low temperature superconductingtoroidal magnets. These are to be used on the space station as part of thehigh energy physics instrumentation used in cosmic ray spectrometry. Thecurrent in the superconducting magnets is about 800 A. They are cooled by4,000 liters of superfluid He at 1.5 K. A major problem involvesdisconnecting the high current Cu leads to the magnet after the persistentcurrent has been started; the possibility of quenching of the supercurrentis a major reliability issue.

3 If the leads to the low temperature magnet contained a high current STIDit should be possible to leave them connected during operation, so thatquenching and explosive eneroy release would be much less likely. Such asystem could be tested in the laboratory in the near future.

3 5.4 Simplicity and Convenience

Finally, the STID is a simple device. In present approaches to thermalisolation careful design- is required. Thin delicate wires are often used,and these may even be tapered. This can lead to difficult installation and a5mechanically weak system of kludges. By contrast a STID would be slmostan off-the-shelf item, would be nearly as simple to install as a set ofresistors, could be easily retrofitted, is mechanically robust, and shouldbe very dependable.

35U£ 15

I

U

ILMSC P008929

3 6. STID DESIGN AND PROTOTYPE PROPERTIES

36.1 Design and Fabrication

The conceptual design of a STID is shown in figure 2. Realization of such aI device requires a substrate with superconducting traces and a housing forthe substrate. The housing must be mechanically strong, have high thermalimpedance, and provide for electrical connections to the superconductingtraces.

5A preliminary housing design used a specially constructed fiberglassframe with low thermal conductivity connecting posts. A clamparrangement secured the substrate to the housing and allowed smallamounts of independent motion of the substrate within the housing forstrain relief. Eight-pin connectors were provided at each end for3 attachment of thin connecting wires from the traces. The preliminaryhousing was rather bulky, about 5 cm by 5 cm by 1 cm, and the connecting

3wires were about 2 cm long.

While the preliminary housing served its purpose, it was difficult andI expensive to construct and the long thin connecting wires were subject to

damage. An improved version was therefore designed, based on experiencewith the preliminary housing. The improved version, the prototype design,was constructed from commercially available integrated circuit holderscalled flatpacks, was simple and cheap to construct, was only about 3 cm3by 3cm by .5cm, and the thin wire interconnects were only about 2cmlong. Details and a photo of the prototype design are given below. Thepreliminary housing was only used for preliminary proof-of-conceptmeasurements of thermal conductivity, while the prototype design wasused for the definitive thermal and electrical testing as discussed below.

Figure 5 shows the prototype STID design. Ten of these were actually builtand screened by electrical testing. Only the five best were tested indetail, and are discussed below. The other five were stored. The heart ofthe STID is the substrate holding the superconducting traces of TIBaCaCuO

1!1

LMSC P008929

(TBCCO). The innermost 22 traces, alternating wide and narrow, are theones actually used. These are nominally 0.8 pm thick and 500 pm or 200pm wide. Gold pads, about 0.5 pm thick, are deposited on the ends of thetraces, to which 0.025 mm gold wires are attached by ultrasonic bonding.The other ends of the gold wires are ultrasonically bonded to the pads ofthe housing.

The superconducting traces of TBCCO are intended to carry at least 5,000A/cm 2 . Thus the 200 pm traces should carry about 8 mA and the 500 pmtraces should carry about 20 mA.

The substrate is polycrystalline ZrO 2, 2.54 cm square by 0.15 mm thick.This material was chosen to provide a good surface for deposition of thetraces, for low thermal conductance, and for strength. (As we will discussbelow it does not appear to actually provide a good surface fordeposition.) Note that the length of the substrate between the housingends is somewhat reduced from 2.54 cm and is only 1.59 cm.

The housing is constructed by sawing the ends from a conventionalflatpack designed for 1 inch square IC chips, then reconnecting them with2 slabs of G10 fiberglass, about .5 cm by .1 mm thick. The thermalconductance of the fiberglass is quite low. The housing has a total of 22leads on each side, as evident in the figure. The substrate with traces isattached to the housing at each end with epoxy. The substrate epoxy jointwas tested by dipping the assembly from room temperature into liquidnitrogen at 77 K.

The device could be wrapped with aluminized mylar for thermal radiationshielding if desired. The overall dimensions, excluding the leads, are about3.2 cm square by .5 cm thick.

17

I LMSC P008929

2

Fiue53h MC TDpooyeI ___ ___18

!

LMSC P008929i6.2 STID Thermal Properties

Thermal testing has been performed on the preliminary housing and oneprototype STID mockup, and electrical testing on all of the 5 prototypeSTIDs

The thermal conductance of a STID can be estimated theoretically fromthe dimensions and the measured thermal conductivity of its materials.The main thermal leakage path is through the ZrO 2 substrate, which has anaverage thermal conductivity of about 15 mW/cm K in the range 10 to 100K. The effective size of the substrate is 2.54 cm by 1.59 cm by 0.015 cmthick; in the test apparatus it sits between temperatures of 77.5 K and 6K. From these numbers and equation [2] the heat leakage can be estimatedto be

P(Z) = 25.7 mW (leakage through zirconia substrate) [6]

The G10 fiberglass slabs leak very little heat since their thermalconductivity is only about 1.8 mW/cm K. For the 2 slabs of G10 which are1.59 cm X .5 cm X .02 cm the heat leakage can be estimated as:

P(F) = 1.6 mW (leakage through fiberglass slabs) [7]

Thus the estimated total heat leakage is:

P(T) = 27.3 mW (total leakage by conduction) [8]

This should be considered as a rough upper limit since we have notincluded effects such as the thermal resistance of the epoxy jointbetween the substrate and the housing and the 2 dimensionalnonuniformity of heat flow in the substrate.

The heat leakage of the device was measured as follows: one side wasmaintained at 6 K, and the other side was connected to a resistance

19

I LMSC P008929

3 heater. The power to the heater was then adjusted to produce atemperature of 77.5 K at the hot side, and the power level measured. The3 measured heat flow, after correction for radiation, was found to be:

P(M) = 24.5 ± 1 mW (measured leakage) [9]

This is in quite good agreement with the above theoretical estimate, sowe may conclude that the thermal operation of the device is wellunderstood. Table 3 summarizes the thermal conductivity results.

3 Table 3: STID Thermal Properties

I Component Area Length Conductivity AT Power

(cm 2) (cm) (mW/cm K) (K) (mW)

Substrate .038 1.59 15 71.5 25.7 (theory)

I Fiberglass .020 1.59 1.8 71.5 1.6 (theory)

3 Total Device .........- 71.5 27 (theory)

3 Measured .........- 71.5 24.5 ± 1

2IIII

! 20

I

I

ILMSC P008929

I6.3 STID Electrical Properties

I The electrical properties of one of the STIDs (2L505.1) are shown inFigures 6 and 7. Figure 6 shows the measured resistance as a function ofT for 6 of the traces. Note that the resistance of the gold connectingwires and other peripheral connections external to the traces is included,so the total resistance does not go to exactly zero. The resistance beginsto rise at about 70 K, with the resistance of one of the traces remainingbelow 1 ohm up to about 78 K. The width of the transition is rather large,about 30 K. Figures 7a to 7f show the resistance as a function of currentfor all 22 of the connected traces at 40 K, 60 K and 80 K. In the 80 K curvethe voltage begins to rise rapidly above about 1 ma (ref. 6). Thiscorresponds to a current density of about 250 A /cm 2 , which is about anorder of magnitude less than anticipated. In Appendix B similar curves forthe other four devices are shown.

Note that there is a great deal of variation in the curves in both figures,with the resistance of the traces varying by about an order of magnitude.Since half of the traces are 500 pm and half are 200 pm some variation isto be expected. Variation between devices is even larger. Some displayuseful operating temperatures as low as 60 K, while the graphs for2L505.1 show about 78 K. The variation of critical current (at which the3 voltage rises abruptly) is also very large.

6.4 STID Properties: Conclusions

The heat leakage allowed by the STIDs is quite low and in good agreementI with theoretical estimates.

The current carrying ability however is over an order of magnitude lower3 than expected: whereas we expected at least 5,000 A/cm 2 when designingthe device we found about 250 A/cm 2 when they were measured. In5 addition the expected T, was about 100 K, whereas the measured values

3| 21

I

THIS

PAGE

IS

MISSING

IN

ORIGINAL

DOCUMENT

1 LMSC P008929

32L505-1 (T=40K) ODD # -500Lufl tracesEVEN # -2oaum traces

U 1000

E__

* __0--- Ri---- R2

-a-- R3

-C'----R6

cn R7

....... ....... ............. .. . . . . . .........Il1.01 1 1 0

1 1000

0 100 R12----- R13-U--- R14

__ __ __---- R15

c) Ri17-*----R18

1 -u----R20

.R2

I .01 .1110

3 Current (mA)

Figure 7a,b. Resistance of STID as function of current at 40 K.

23

I

I LMSC P008929

I 2L505.1 (T=60K) ODD # 500um tracesEVEN # 200um traces

I 1000

C 100 --- Ri- - R2

* ~ 1 _ _ _ 1 1-6-- R3,-. R410 .......... .. ........ R5

0 -C- R6R7

- - R8

1 10 -- Ru

10 00

E1 100 R12

- - R13-U--- R14

10__ - R15o 10 . R16

- - R17.... ... .... R 1 8

I& Rig8i -f -- R19- ---- R20I-------- R21

IR22

I .01 .1 1 10

3 Current (mA)

Figure 7c,d. Resistance of STID as function of current at 60 K.

I| 24

I

I

i LMSC P008929

i 2L505.1 (T=80K) ODD # - 500um tracesEVEN # 200um traces

1000

E

o 100 - -- Ri

- - R2-U-- R3

__________ - ~ -- R4cc 10 ............... R 5

R7

.01 .1 1 10'O~ -

IIR7

00

E

0 100 -.--. R12

- R13

I i R15

0----- R16

00 R18cc-- R19

1 --- '--- R20-+------ R21~- R22

1 01 .11 10

Current (mA)

Figure 7e,f. Resistance of STID as function of current at 80 K.

iI ! I I25

I

1 LMSC P008929

I varied between about 60 K and 80 K.

3 Thus the prototype STIDs function qualitatively as desired and are usefulfor testing. But they are not practical devices due to the low currentcapacity, and to a lesser extent due to the low transition temperature.The reasons for this will be discussed further in the next sections.

2IIIIIIIIIIU

| 26

I

I

I LMSC P008929

3 7. PROTOTYPE PROBLEMS AND SOLUTIONS

i 7.1 Design and Fabrication

Many minor problems were encountered in the design and fabrication ofn the STIDs, but all were eventually solved. Table 4 gives a summary of

some of them.

i Table 4: Design and Fabrication Problems and Solutions

Problem Solution

Select thermally insulating Zirconia (ZrO 2) strong and lowsubstrate material conductivity

U Obtain thin substrate Zirconia ground down to .15 mm

(6 mil) by STI

I Deposit long TBCCO traces 1 inch traces produced by STI

3 Design simple strong thermal Cut ends from flatpackinsulating housing Fiberglass side strips

I (see Figure 6) Epoxy adhesive

Bond connecting wires Pairs of 1 mil gold wire3 (see Figure 7) Ultrasonic wedge bonderCareful adjustmentExpert operator

The wire bonding problem was the most chronic and difficult. Thermalcompression bonding was tried, but the heat often destroyed thesuperconductor, Ultrasonic bonding often failed and sometimes lifted thegold layer. The problem was solved by testing different wire materialsand sizes and bonding techniques. We finally settled on thin (1 mil) soft

27

I

I LMSC P008929

I gold wire in a carefully adjusted ultrasonic wedge bonder operated by anexperienced expert operator. Figure 8 shows a typical wire bond made

* after the problem was solved.

7.2 Basic Physics

Two basic physics problems were encountered with the superconductingtraces, as noted in section 6.4. The lesser problem was that the transitiontemperature was about 78 K or lower instead of the 100 K expected. Oneobvious solution is to select the STIDs with the best electrical propertiesout of those avai!able. Appendix B shows the large variation in theelectrical properties of the STIDs.

I The major problem arose as noted in section 6.4, that the current capacityof the traces was over an order of magnitude less than expected. Thisprevents the STIDs from being practical, although they functionqualitatively correctly and can be used for testing. This problem iscertainly due to the poor morphology of the TBCCO traces onpolycrystalline ZrO 2 , and is contrary to our expectation that

polycrystalline ZrO 2 should be a good substrate. Figure 9 shows a SEMmicrograph of the traces; the jumbled nature of the crystals shows thatconduction must be via a torturous path of small cross sectional areathrough the logjam! Figure 10 shows TBCCO crystals on a substrate ofLaAIO3, with greater smoothness and uniformity evident.

* The trace quality problem appears to be due to a fundamentalincompatibility of the ZrO 2 substrate with TBCCO. In the next sections we

* will discuss this further.

IIII

U LMSC P008929

0

'AA 0

0

__ 0

0

C6

ER,)

290

LMSG P008929

cJ

0

c0

300

LMSC P008929

I-COZO AILr.37

007 2KIjjO 08

Fiue1.ICOtae nZ0 to)cmae oLA0 bto)I3

LMSC P008929

8. MARK I TEST DEVICE

8.1 Intent

The next phase of the project was to design and fabricate 5 Mark I test

devices with TBCCO traces on a LaAIO 3 substrate. The intent was to see

what current capacity could be obtained in TBCCO traces. Since LaAIO 3 is a

poor thermal insulator, evident from the thermal conductivity numbers inFigure 3, the devices were not intended to be thermal isolators.

8.2 Design and Fabrication

The design and fabrication of the Mark I test device are very similar to theSTID. Superconducting traces of TBCCO are deposited on LaAIO 3 substrates

by STI. These are nominally .8 pm thick and alternate 500 pm and 200 pmwide.

The expectation was that the traces should have a transition temperature

of about 100 K and a current capacity in excess of about 10,000 A/cm2

The actual values obtained equal or exceed these numbers.

The substrates are single crystals of 1 cm square LaAIO 3, about half the

size used in the STID. Substrate and traces are epoxied into a smallstandard flatpack with 11 leads. Electrical connections are made withgold wires as in the STID.

We emphasize that the test device is not intended or designed to bethermally isolating, since it is only an electrical test device. Five suchtest devices have been built and tested.

8.3 Electrical Properties

The electrical properties of the test device are extremely good, especially

when compared with the STID. Figure 11 shows the resistance as a

32

I

LMSC P008929

I function of temperature for test device 2L655.1; 4 leads are shown. Thetransition to superconducting at about 102 K is evident, and is only a few

5 K wide. This is in agreement with the STI measurement usingsusceptibility, and is consistent with a large fraction of the bulk of thematerial being superconducting. Note also the low value of about .2 ohmfor the residual resistance of connecting wires etc. It is impressive tocompare Figure 11 for the Mark I test device with Figure 6 for the STID!

I Figure 12 shows resistance as a function of current for the same test

device. Figure 12a for 80 K indicates no sign of increasing resistance upto a current of 600 mA, corresponding to a current density of about375,000 A/cm 2 in the 200 pm traces. The current capacity of the tracesmay be well above this, but we did not extend the measurements to highercurrent since the external connections are not expected to carry much

i more.

Figure 12b for 100 K shows a clear rise in resistance at about 100 mA forI the 200 pm traces and about 300 mA for the 500 pm traces,

corresponding to current capacity of about 60,000 A/cm 2 . This also isremarkably large considering how close the temperature is to thetransition temperature.

Some of the resistance measurements in Figure 12 were made using ashort pulse technique instead of DC, due to the large currents involved.Figure 12a shows a small glitch at 20 mA where the pulse measurements

3IIUI 33

I

I

LMSC P008929

l

3 2L655.1

ODD # 200um tracesEVEN # -500um traces

a2 10E

0 1 #5• -- .e. #6e -- --a---- #7u #8

cc

.190 100 110 120

Temperature(K)

Figure 11. Resistance of M 1 test device as function of temperature

34

LMSC P008929

2L655.1 (T=8OK) ODD # - 500um tracesEVEN # - 200um traces

S10

E

o00-- #1

_- #2U* -U #3

C13 ........... . . .. #5"6------ #4

- #6

,* #7-a-- #8rr -U- #9

4-- #10

#11

10 100 1000

Current (mA)

Figure 1 2a. Resistance of M 1 test device as function of current at 80 K

35

LMSC P008929

I2L655.1 (T=100OK) ODD # -500um tracesEVEN # 200um traces

2#

-U---#3

1 ___ _4

£~I 1 #1000 10

3 Current (mA)

Figure 12b. Resistance of M 1 test device as function of current at 100 K

36

I

LMSC P008929

I begin, and shows that the 2 measurement techniques are reasonablyconsistent.

Figure 10 is an SEM photo of the surface morphology of the TBCCO tracesI in the test device and also in a STID. Although rather irregular the

structure of TBCCO on LaAIO 3 is much smoother than the TBCCO on ZrO 2.

8.4 Mark I Test Device Properties: Conclusions

We conclude that the electrical properties of the test device areextremely good; the transition temperature equals that desired andexpected, and the current capacity exceeds that expected by over an orderI of magnitude. Compared to the STID the transition temperature is over 20K higher, and the current capacity is about 3 orders of magnitude greater.

3IIII£III1 37

U

!

LMSC P008929I9. MARK II TEST DEVICE

£ 9.1 Intent

I The final phase of the project was to design and fabricate two testdevices, termed Mark II, with TBCCO traces, one on single crystal ZrO 2 and

j one on polycrystalline ZrO 2 The intent was to find the reason for the lowcurrent capacity in the prototype STIDs, which all used polycrystallinematerial. Like the Mark I test devices the Mark Ils were not intended tobe thermal isolators but strictly electrical test devices, specifically todetermine if the current capacity problem was due to the use ofpolycrystalline substrate.

1 9.2 Design and Fabrication

The Mark II test devices use TBCCO traces of widths 200 .tm and 500 itmtike the prototype STIDs and the Mark I test devices. One Mark 11 deviceuses a 1 cm square substrate of polycrystalline ZrO 2 like the prototypes,

and the other uses a 1 cm square substrate of single crystal ZrO 2

Otherwise, construction is exactly like the Mark I test devices.

5 9.3 Electrical Properties

The current capacity of the traces on the polycrystalline substrate is onlyabout 25 A/cm2 , even lower than for the prototype STIDs; moreover the

transition to superconductivity is quite wide, about 20 K or more.

The current capacity of the traces on the single crystal substrate is overtwo orders of magnitude larger, about 5,000 A/cm2 , which is adequate forpractical use.; moreover the transition to superconductivity is at about100 K and very narrow.

Figure 13 shows the resistance as a function of temperature for both MarkII devices, 2L759.1 being single crystal and 2L759.3 being polycrystalline.

38

II

!

LMSC P008929

IThe transition to superconcuctivity begins at about 100 K for all tracesshown, but is very much narrower and goes much lower for the singlecrystal device. In particular traces number 9 and 11 are especially good.

Figures 14a and 14b show resistance versus current for both devices at 77K. Traces on the single crystal device carry about 20 mA before theresistance begins to rise, which corresponds to a current density of about5,000 A/cm 2 . By contrast traces on the polycrystalline device carry only

about 0.1 mA, corresponding to only about 25 A.cm 2 .

I One serious problem was discovered with the single crystal device. Overthe time period 5/17/91 to 6/25/91 the device degraded significantly.

I! Figure 15 shows this clearly. Note in particular the increase in resistancefor the best trace, number 9, at 30 mA; it increased from about 0.11 ohmto about .21 ohm. This indicates that a passivation coating is probablynecessary

5I 9.5 Mark 11 Test Device Properties: Conclusions

In summary it is clear that the poor electrical properties of the prototypeSTIDs are due to the polycrystalline substrates, and can be corrected withthe use of a single crystal substrate. This is a straightforward if

5 somewhat expensive solution. The degradation over time may becorrectable with the use of a passivation coating.

3

I

1 39

U

LMSC P008929

00

C.) L- .02

c 0 >

in CD-

o 1o

CCu

=S I0 3U1IO

>~ Cu

3l c LMSC P008929

CC.

C.))

518 8 - . -0 E

5S '0 0 - L.i.U)

41,

£ LMSC P008929

CYa)

£ ~C

I _ _A

(4-

3QU)

_ _ _ _ _ _ _ _ _ .Cc

(SLU4 ) GOUISI0)

I 2Z42

THIS

PAGE

IS

MISSING

IN

ORIGINAL

DOCUMENT

I5 LMSC P008929

1 1. INTRODUCTION (PART II)

Passive microwave components made of patterned superconducting films onI dielectric substrates have low resistance and generally high performance; they

are thus of great interest. The patterning of such films is typically done by wetIetching, but this entails a fundamental problem; considerable undercutting oftenresults, leaving irregular edges in the pattern. One possible solution is tocorrect the undercutting with the use of a distorted initial profile. This,

I however, is not a very robust manufacturing approach, so other solutions arepreferred.

I The reason that wet etching undercuts TIBaCaCuO (TBCCO) films so badly is thatthe nominally isotropic etch in fact prefers the 'a' and 'b' directions to the 'c'Sdirection. The film is oriented with the 'c' direction normal to the substrate,hence the undercutting. The basic task addressed in this study is to evaluate dryetching or ion milling as an alternative to wet etching in order to avoid theundercutting.

I Ion milling is intrinsically highly anisotropic, and is known therefore topreserve lateral film dimensions very accurate!y. Thus there should be verylittle undercutting and distortion of thin film traces. However it is also knownthat ion milling produces a temperature rise that may degrade thesuperconducting properties of the TBCCO film. The solution explored in this

I prcject is to use active substrate cooling to protect the superconducting tracesfrom the temperature rise.

4III3

| 45

I

LMSC P008929

2. TECHNICAL ACHIEVEMENTS

Argon milling should be capable of dry etching fine geometries in thin filmswith virtually no pattern undercutting, provided of course that the photoresistmask remains intact during etching. The main problem with the procedure isheating and ion bombardment damage to the film, which can degrade itssuperconducting properties.

We have reevaluated our original milling equipment, with regard to its effect onsample heating, and have changed to a 20 cm Technics Corp. mill which has muchmore efficient sample cooling at higher beam current densities. A recentlydeposited TBCCO sample (2L788.4) was coated with 1.5 p.m of positive resistand patterned with the 6 mil resonator geometry from STI mask #23. It wasthen argon ion milled at 500 V with a current density of 0.32 mA/cm 2 . Thesample temperature did not exceed 600 C during the milling process asdetermined by an on-chip temperature monitor. Since the temperature remained

well below the nominal upper bound of 1200 C future milling may use a highercurrent density to increase the milling rate. The ion milling rate of the TBCCOfilm under these conditions was estimated to be about 100 A/min. The millingtime was extended to insure complete film removal in the area outside the 6mil pattern. No residue of copper oxide remained after ion milling as has beenobserved with wet etching. The remaining pattern thickness was measured at0.69 Iim. The pattern width was measured as 152 Iim versus a nominal 153 pimdesign value. Thus the expected accuracy was achieved.

After processing at LMSC the sample was returned to STI for resonatormeasurements. These indicated that there had been some degradation of themicrowave characteristics during processing. It appears that the sample mayhave degraded due to exposure to the air tor an extended time between processsteps. Such exposure can easily be minimized in the future. The photoresistthickness should also be increased to eliminate actual contact between the ionbeam and the active sample, while still permitting some over-milling to clearthe unmasked area. This over-milling can be quantitatively verified by initiallymeasuring the exact film thickness with a profilometer before attempting anyactive area processing.

46

I

U LMSC P008929

3 APPENDIXA: THERMAL CONDUCTIVITY AVERAGES

In the basic thermal power equation [2] the appropriate value of thethermal conductivity to use is the average over temperature. This isintuitively reasonable and easy to demonstrate.

Consider a short segment of thermally conducting wire of length dx andarea A, with a temperature difference of dT. The definition of thermalconductivity is

P = Kth(T) A dT/dx [a-i]

Power is understood to flow down the temperature gradient. Integratethis over the length of the wire, L , over which the temperature variesfrom Ti to Tf, to obtain

Tf3P L = A f th(T) dT = A <Kth> AT [a-2]Ti

I where AT = Tf - T i . Thus:

SP = A <th>AT/L. [a-3]

3 This is the basic power equation used in the text, with <Kth> denoted

simply by Kth.

AI

! A-i

I

LMSC P008929

APPENDIX B: ELECTRICAL DATA FOR PROTOTYPE STIDs

2L505.1 2L504.110000 10000 .

~1000 1000

E-- 3- # ----- V4

C 100 f --- ',-,-'--''" #54 100 -f , --- :8

40 60 80 100 120 140 40 60 80 100 120 140

2 2L 515. 1 2 L 512. 1

10000 10000

_1000 1000

E 0- ~#5 a - 1o100 a--:16 100 -a- 2-- - #7 . 0-- 3

-s0- #1 Z5o 1

40 60 80 100 120 140 40 60 80 100 120 140

Temperature (K)

2L506.110000

1000

E

O 100 -- - #4

S 10 ---- #6C 10 - - --

1

40 60 80 100 120 140

Temperature(K)

Figure b-1. Resistance of STIDs as function of temperature.

B-1

LMSC P008929

2L505.1 (T=80K) ODD # 500um tracesEVEN # -2oaum traces

0 100 - - Ri

---- R2----0---R3

S10 1 i:gI _____ ___R5

-U---R6.- ..... R i7

10000

00

EI0 100 ~-s-- R123 -.---- R13

0 Ri18

11----- R20-I---- R213 -~-R22

.01 .1110

Current (mA)

Figure b-2. Resistance of STID 2L505.1 as function of current at 80 K.

B-2

LMSC P008929

2L5 15.1 (T=80K) ODD #-500um traces3 EVEN # -200um traces

0 00 __ _ _

E* ~ 100 - -- ---

0 R_ _ _ _ _ -S --- R2

_______R4

- R5-a-- R6

00_ _ _ _ _ _ _ _ _ _ R7-cc- R8

1 -u----R9-4 - RioI -~-- Ri 1

1.01 .1 1 10

1000 1'

-0- R12

____ _ __ ___ U--- R14_____ ____ _ __ _____ __ -~ --- R15

-0---- R17

Rig81R213 ----s---R22

.01 .1 1 10

Current (mA)

Figure b-3. Resistance of STID 2L51 5.1 as function of current at 80 K.

B-3

LMSC P008929

2L506.1 (T=80K) ODD # - 500um tracesEVEN # 200um traces

1000 . . . . . . . ... . .

E10 ------- Ri0" G R1

- R2- U - R3_ __ _- R4

S10R5R6

0 .R7n- R8

1 --- R9

4- R10Ril

.01 .1 110

1000

E100O -0-- R12

....- - R 13--- -- R14-- ,- R15

m 10 ".......... . R 16

-- ----- R17• ...... R18m Rig8-0 - R19

1 . ..-- R20R21

a---- R22

.0110

Current (mA)

Figure b-4. Resistance of STID 2L506.1 as function of current at 80 K.

B-4

3 LMSC P008929

2L504.1 (T=80K) ODD # -500um tracesEVEN #200um traces

00

E=~ 100o0a-- RI-.---- R2

* -a---R3

_ _ _ _ __ _ _ ~R410--- R53 -si----R6

3 01 1 1R1

I ___R8

30 1000

-1--- R12-.--- R13----- R14

* - * -- R15

-D-- R17

0 _ _ _ _ _ _ _ _ _ _ _ _ _ _ * -- R19~ 1 ----- R20

-1----R213 -s----R22

1 01 .1 1 10

Current (mA)

Figure b-5. Resistance of STID 2L504.1 as function of current at 80 K.

B-5

LMSC P008929

2L512.1 (T=80K) ODD U 500um tracesU EVEN # -200um triaces

U ~100

0 R

N R3

1004

+U - Rio

100Ri0 '-0-- R12

cc 1 Ri16-a----R17

1 ~----- R20-*---- R21-s-- R22

.01 .1 1 10

Current (m A)

Figure b-6. Resistance of STID 2L51 2.1 as function of current at 80 K.

B-6

I

LMSC P008929

APPENDIX C: ELECTRICAL DATA FOR MARK I TEST DEVICE

2- 2L651.3 2L651.4

100 0 1 0

; 10 10 -o- - z4

o 1. . - ...-- #. . . . . . . .#

E -0 - 3

*1

.1 17

g0 i00 110 120 V0 100 110 120

2 L6 54.4 2 L6 55. 1

100 100

a6 100- #

#7 .1

80 100 110 120 go 100 110 120

Temperature (K)

2L655.4100

C-10Cal

90 100 110 120

Temtperature (K)

Figure c-i . Resistance of M I test devices as function of temperature.

C-i

LMSC P008929

2L655-1 (T~l1OOK) ODD I - o0um tracesEVEN # -200um traces

10

E

0~ #2

* -*--- #30 - 4 #4

1 #5

o #7-*- #8

-s-----#11

.#1

1 10 100 1000

Current (mA)

Figure c-2. Resistance of M I test device 2L655-1 vs. current at 100 K

C-2

1 LMSC P008929

32L651.4 (T=100OK) ODD # 500um tracesEVEN # 200um traces

U ~10 ___

E

-0-

I #7-a-- #8

3 #101 -4----#11

31 10 100 1000

Current (mA)

Figure c-3. Resistance of M 1 test device 2L651.4 vs. current at 100 K

C-3

LMSC P008929

2L654.4 (T=1 00K) ODD # 500um tracesEVEN # 200um traces

10 * . --

E

0~ ca"u 1'

1 -. 3---Cokxm H

AW ....... t ...... Colum ItCr.-A Cokrm 2(

= Coum 21SCabn 2Z

-6- Coum z~

10100 1000

Current (mA)

Figure c-4. Resistance of M I test device 2L654.4 vs. current at 100 K

C-4

LMSC P008929

2L655.4 (T=1 00K) ODD a M O~m tracsEVEN # 200w. Imaes

10

E

* 83 (01=4~C __ _ __ _ __ _ __ _ - *Ouu

* 0 1(Ohmns

A 1 7 T.I I Ohm

1- 10 100 1000

Current (mA)

Figure c-5. Resistance of M I test device 2L655.4 vs. current at 100 K

C-5

LMSC P008929

2L651..3 (T=I100K) ODD # 500um tracesEVEN # 200um traces

10

o -'--- S3

0 84I 86

-U-- 810

1 0100 1000

Curront (mA)

Figure c-6. Resistance of M I test device 2L651.3 vs. current at 100 K.

C-6