technology focus computers/electronics - nasa · technology focus computers/electronics software...

Technology Focus

Computers/Electronics

Software

Materials

Mechanics

Machinery/Automation

Manufacturing

Bio-Medical

Physical Sciences

Information Sciences

Books and Reports

04-03 April 2003

https://ntrs.nasa.gov/search.jsp?R=20110023770 2020-06-12T02:09:25+00:00Z

INTRODUCTIONTech Briefs are short announcements of innovations originating from research and develop-

ment activities of the National Aeronautics and Space Administration. They emphasizeinformation considered likely to be transferable across industrial, regional, or disciplinary linesand are issued to encourage commercial application.

Availability of NASA Tech Briefs and TSPsRequests for individual Tech Briefs or for Technical Support Packages (TSPs) announced herein shouldbe addressed to

National Technology Transfer CenterTelephone No. (800) 678-6882 or via World WIde Web at www2.nttc.edu/leads/

Please reference the control numbers appearing at the end of each Tech Brief. Information on NASA’s Commercial Technology Team, its documents, and services is also available at the same facility or on the World Wide Web at www.nctn.hq.nasa.gov.

Commercial Technology Offices and Patent Counsels are located at NASA field centers to providetechnology-transfer access to industrial users. Inquiries can be made by contacting NASA field centersand program offices listed below.

Ames Research CenterCarolina Blake(650) [email protected]

Dryden Flight Research CenterJenny Baer-Riedhart(661) [email protected]

Goddard Space Flight CenterNona Cheeks(301) [email protected]

Jet Propulsion LaboratoryArt Murphy, Jr.(818) [email protected]

Johnson Space CenterCharlene E. Gilbert(281) [email protected]

Kennedy Space CenterJim Aliberti(321) [email protected]

Langley Research CenterSam Morello(757) [email protected]

John H. Glenn Research Center atLewis FieldLarry Viterna(216) [email protected]

Marshall Space Flight CenterVernotto McMillan(256) [email protected]

Stennis Space CenterRobert Bruce(228) [email protected]

Carl RaySmall Business InnovationResearch Program (SBIR) &Small Business TechnologyTransfer Program (STTR)(202) 358-4652 [email protected]

Dr. Robert NorwoodOffice of CommercialTechnology (Code RW)(202) 358-2320 [email protected]

John MankinsOffice of Space Flight (Code MP)(202) 358-4659 [email protected]

Terry HertzOffice of Aero-SpaceTechnology (Code RS)(202) 358-4636 [email protected]

Glen MucklowOffice of Space Sciences(Code SM)(202) 358-2235 [email protected]

Roger CrouchOffice of Microgravity ScienceApplications (Code U)(202) 358-0689 [email protected]

Granville PaulesOffice of Mission to Planet Earth(Code Y) (202) 358-0706 [email protected]

NASA Field Centers and Program Offices

NASA Program Offices

At NASA Headquarters there are seven major program offices that develop and oversee technology projects of potential interest to industry:

NASA Tech Briefs, April 2003 1

5 Technology Focus: Fastening/Joining/Assembly Technologies

5 Tool for Bending a Metal Tube Precisely in a Confined Space

6 Multiple-Use Mechanisms for Attachment to SeatTracks

7 Force-Measuring Clamps

7 Cellular Pressure-Actuated Joint

9 Computers/Electronics

9 Block QCA Fault-Tolerant Logic Gates

10 Hybrid VLSI/QCA Architecture for Computing FFTs

11 Arrays of Carbon Nanotubes as RF Filters inWaveguides

12 Carbon Nanotubes as Resonators for RFSpectrum Analyzers

15 Software

15 Software for Viewing Landsat Mosaic Images

15 Updated Integrated Mission Program

15 Software for Sharing and Management ofInformation

15 Update on Integrated Optical Design Analyzer

17 Materials

17 Optical-Quality Thin Polymer Membranes

19 Mechanics

19 Rollable Thin Shell Composite-MaterialParaboloidal Mirrors

19 Folded Resonant Horns for Power UltrasonicApplications

21 Machinery/Automation

21 Touchdown Ball-Bearing System for MagneticBearings

22 Flux-Based Deadbeat Control of Induction-MotorTorque

23 Manufacturing

23 Block Copolymers as Templates for Arrays ofCarbon Nanotubes

25 Physical Sciences

25 Throttling Cryogen Boiloff To Control Cryostat Temperature

27 Information Sciences

27 Collaborative Software Development ApproachUsed to Deliver the New Shuttle TelemetryGround Station

29 Books and Reports

29 Turbulence in Supercritical O2/H2 and C7H16/N2Mixing Layers

29 Time-Resolved Measurements in Optoelectronic Microbioanalysis

04-03 April 2003

This document was prepared under the sponsorship of the National Aeronautics and Space Administration. Neither the United States Governmentnor any person acting on behalf of the United States Government assumes any liability resulting from the use of the information contained in thisdocument, or warrants that such use will be free from privately owned rights.



A relatively simple, manually oper-ated tool enables precise bending (typ-ically, within ±1/2° of the specifiedbend angle) of a metal tube located ina confined space, with a minimum offlattening of the tube and without sig-nificant gouging of the tube surface.The tool is designed for use in a situa-tion in which the tube cannot be re-moved from the confined space forplacement in a conventional bench-mounted tube bender. The tool is alsodesigned for use in a situation in whichpreviously available hand-held tube

benders do not afford the requiredprecision, do not support the tube wallsufficiently to prevent flattening orgouging, and/or do not fit within theconfined space.

The tool is designed and fabricatedfor the specific outer diameter and bendradius of the tube to be bent. The tool(see figure) includes a clamping/radiusblock and a top clamping block that con-tain mating straight channels of semicir-cular cross section that fit snugly aroundthe tube. The mating portions of theclamping/radius block and the top

clamping block are clamped around alength of the tube that is adjacent to thebend and that is intended to remainstraight. The clamping/radius block isso named because beyond the straightclamping section, its semicircular chan-nel extends to a non-clamping sectionthat is curved at the specified bend ra-dius. A pivot hole is located in theclamping/radius block at the center ofthe bend circle.

The tool includes a bending blockthat, like the other blocks, contains astraight semicircular channel that fits

Technology Focus: Fastening/Joining/Assembly Technologies

A Tube Is Clamped and Bent in contact with conformal tubelike surfaces that provide full support with minimal damage.

Pivot PinArms

Top Clamping Block

Top Clamping Block

Clamping/Radius Block

Clamping/Radius Block

Channel Curved to Bend Radius

Pivot Center Line

Tube

ThreadedRod

SIDE VIEW OF TOOL AND TUBE BEFORE BENDING

SIDE VIEW OF TOOL AND TUBE ATCOMPLETION OF BEND

AXIAL VIEW OF CLAMPING/RADIUS BLOCK AND TOP CLAMPINGBLOCK CLAMPED AROUND TUBE BEFORE BENDING

Tool for Bending a Metal Tube Precisely in a Confined SpaceThis tool offers capabilities that prior tools do not.Goddard Space Flight Center, Greenbelt, Maryland

6 NASA Tech Briefs, April 2003

around the outside of the tube. Thebending block contains a pivot hole tobe aligned with the pivot hole in theclamping/radius block. Once the tubehas been clamped between the clamp-ing/radius and top clamping blocks,the bending block is placed around thetube, the pivot holes are aligned, and a pivot pin is inserted through the pivotholes.

To bend the tube, the bending blockis pivoted so that its semicirculargroove slides along the tube, forcingthe tube into the curved portion of the

groove in the clamping/radius block.An arm that extends from the clamp-ing/radius block and a similar arm thatextends from the bending block pro-vide mechanical advantage for generat-ing bending torques and forces. Thesearms are actuated by turning a nut on athreaded rod that runs through holesin both arms.

To ensure a precise bend, one shouldmeasure the bend angle by use of a pro-tractor at intervals during the bendingoperation. Even so, it is desirable to cal-ibrate the tool in two ways: (1) measur-

ing and/or calculating the increase inthe bend angle for each turn of the nutand (2) measuring and/or calculatingthe amount of springback. Calibrationshould facilitate the approach to thefinal stage of bending (with a slightover-bend to allow for springback) withgreater assurance that at the end, thetube will be bent to the desired anglewithin ±1/2°.

This work was done by Gary T. Davis ofGoddard Space Flight Center. Further in-formation is contained in a TSP (see Page 1).GSC-14412

Multiple-Use Mechanisms for Attachment to Seat TracksThese could serve as standard or universal seat-track clamps.Lyndon B. Johnson Space Center, Houston, Texas

A Seat Track Attach Mechanism (SAM)is a multiple-use clamping device in-tended for use in mounting various ob-jects on the standard seat tracks used onthe International Space Station (ISS). Thebasic SAM design could also be adapted toother settings in which seat tracks areavailable: for example, SAM-likedevices could be used as univer-sal aircraft-seat-track mountingclamps.

A SAM (see figure) is easilyinstalled by inserting it in a seattrack, then actuating a lockinglever to clamp the SAM to thetrack. The SAM includes anover-center locking feature thatprevents premature disengage-ment that could be caused bysome inadvertent movementsof persons or objects in thevicinity.

A SAM can be installed in,or removed from, any posi-tion along a seat track, with-out regard for the locations ofthe circular access holes.Hence, one or more SAM(s)can be used to mount an ob-ject or objects on a track or apair of tracks in an infinitenumber of preferred configu-rations. A SAM can be incor-porated into a dual swivel de-vice, so that two of the SAMscan be made to lock onto twoside-by-side seat tracks simul-taneously, as would be thecase in a standard ISS rack baywhere two side-by-side racksreside. The main benefit to

using two SAMs in a side-by-sidearrangement is to provide a coupledload. By picking up load points on twoseat tracks, a coupled loading is cre-ated, improving the stability andstrength since the load is spread to twoseat tracks at a short distance.

This work was done by Martin Fraskeand Rich May of Johnson Engineering Corp.for Johnson Space Center. For further information, contact the Johnson Commer-cial Technology Office at 281-483-3809;[email protected]

SAM

SAM INSTALLED IN SEAT TRACK

Steel Housing

Steel Paws

EXPLODED VIEW OF SAM

SEAT TRACK

A SAM Is Inserted in, and Clamps Onto, a Seat Track. The SAM can be positioned anywhere along the track, withoutregard for the locations of the access holes.


Force-measuring clamps have been in-vented to facilitate and simplify the taskof measuring the forces or pressures ap-plied to clamped parts. There is a criticalneed to measure clamping forces orpressures in some applications — for ex-ample, while bonding sensors to sub-strates or while clamping any sensitive ordelicate parts. Many manufacturers ofadhesives and sensors recommendclamping at specific pressures whilebonding sensors or during adhesivebonding between parts in general.

In the absence of a force-measuringclamp, measurement of clamping forcecan be cumbersome at best because of theneed for additional load sensors and load-indicating equipment. One prior methodof measuring clamping force involved theuse of load washers or miniature load cellsin combination with external powersources and load-indicating equipment.Calibrated spring clamps have also beenused. Load washers and miniature loadcells constitute additional clamped partsin load paths and can add to the destabi-lizing effects of loading mechanisms.Spring clamps can lose calibration quicklythrough weakening of the springs and arelimited to the maximum forces that thesprings can apply.

The basic principle of a force-measur-ing clamp can be implemented on aclamp of almost any size and can enablemeasurement of a force of almost anymagnitude. No external equipment is

needed because the component(s) fortransducing the clamping force and thecircuitry for supplying power, condition-ing the output of the transducers, and dis-playing the measurement value are allhoused on the clamp. In other words, aforce-measuring clamp is a completeforce-application and force-measurementsystem all in one package. The advantageof unitary packaging of such a system isthat it becomes possible to apply the de-sired clamping force or pressure with pre-cision and ease.

Like many other load-measuring de-vices, a force-measuring clamp containsstrain gauges and exploits the well-knownproportionality between strain and ap-plied force or pressure. More specifi-cally, a force-measuring clamp containsfour strain gauges electrically connectedin a Wheatstone bridge. The bridge out-put is fed to zero and span circuitry, theoutput of which is digitized and dis-played. The span and zero circuitrymake it possible to calibrate the bridgeoutput to indicate force or pressure inany suitable unit of measure.

The strain gauges can be installed byuse of Measurements Group M-Bond 610(or equivalent) epoxy-phenolic adhesiveor Measurements Group AE 10 (or equiv-alent) epoxy adhesive. The strain gaugesare connected to a terminal strip for in-corporation into the bridge by wires of 34American Wire Gauge [≈6.3 mils (≈0.16mm) in diameter]. Wires of the same size

are used for connections between the ter-minal and a printed-circuit board. Theprinted-circuit board contains a voltage-regulation circuit, the span and zero cir-cuits, two watch batteries, and a powerswitch. The final-stage output of theprinted circuit is fed to a digital-display de-vice that is plugged into the printed-cir-cuit board, and is controlled by the zeroand span circuits. Operation of the force-measuring clamp is easy: One simplyslides the power switch to “on,” adjusts thedisplay to zero if necessary, applies aclamping force, and reads the display.

The functionality of a “breadboard”prototype force-measuring clamp wastested in a laboratory by use of the combi-nation of certified weights, a load washer,a strain indicator, and a voltmeter. Somealternate or future embodiments of force-measuring clamps may include smallerbatteries and/or smaller digital displaysfor the sake of compactness, more op-tions, and better packaging of all compo-nents. Force-measuring clamps and/orsimilar devices could also be incorporatedinto other mechanisms.

This work was done by Mark Nunnelee ofDryden Flight Research Center.

This invention has been patented by NASA(patent pending). Inquiries concerning nonex-clusive or exclusive license for its commercial de-velopment should be addressed to Yvonne Kel-logg, Technology Commercialization Specialist,Dryden Flight Research Center, (661)276-3720. Refer to DRC-99-37.

Force-Measuring ClampsClamping forces can be measured easily and quickly.Dryden Flight Research Center, Edwards, California

A modification of a pressure-actuatedjoint has been proposed to improve itspressure actuation in such a manner asto reduce the potential for leakage ofthe pressurizing fluid. The specific jointfor which the modification is proposed isa field joint in a reusable solid-fuelrocket motor (RSRM), in which the pres-surizing fluid is a mixture of hot com-bustion gases. The proposed modifica-tion could also be applicable to otherpressure-actuated joints of similar con-figuration.

The RSRM field joint (see figure)includes a pressure-actuated memberdenoted the J-leg, which is part of abody of insulation. A pressure-sensi-tive adhesive (PSA) is used to bondthe sealing surface of the J-leg to thesealing surface of another body of in-sulation. The pressure actuation issupposed to push these two sealingsurfaces together. However, experi-ence with the joint indicates that itmay not behave as a truly pressure-ac-tuated system.

The modified version of the jointwould be the same as the unmodifiedversion, except that pockets of emptyvolume would be introduced into thebody against which the J-leg is pressed.The size, shape, and orientation of thepockets would be such upon the applica-tion of pressure to the hot-gas side of theJ-leg, the volumes of the pockets wouldnot change significantly and hence thepressures in the pockets, would notchange significantly; the net resultwould be that the pockets would support

Cellular Pressure-Actuated JointPockets in one of the sealing members would help maintain differential pressure.Marshall Space Flight Center, Alabama

the buildup of enough differential pres-sure across the J-leg so that the jointwould behave more nearly like a trulypressure-actuated system.

The breakup of the empty volumeinto multiple pockets would counteractthe tendency toward leakage that wouldbe incurred by introducing a single

larger pocket. In the unlikely event thata gas path penetrated the joint, onlyone pocket would be compromised,while the others would continue tomaintain the differential pressureneeded for pressure actuation.

This work was done by John R. McGuire ofThiokol Propulsion for Marshall SpaceFlight Center.

Title to this invention has been waivedunder the provisions of the National Aero-nautics and Space Act {42 U.S.C. 2457(f)}to Thiokol Propulsion. Inquiries concerninglicenses for its commercial developmentshould be addressed to

Thiokol PropulsionPO Box 707Brigham City, UT 84302-0707Tel. No.: (435) 863-3511Fax No.: (435) 863-2234Refer to MFS-31652, volume and number

of this NASA Tech Briefs issue, and thepage number.


The Modified RSRM Field Joint would be the same as the unmodified one, except for the introduc-tion of the pockets.

Propellant

Propellant

Clevis

Tang

Insulation

Pressurization Slot

J-Leg

Insulation


Computers/Electronics

Block QCA Fault-Tolerant Logic GatesIt may become possible to relax manufacturing tolerances in molecular-scale devices.NASA’s Jet Propulsion Laboratory, Pasadena, California

Suitably patterned arrays (blocks) ofquantum-dot cellular automata (QCA)have been proposed as fault-tolerant uni-versal logic gates. These block QCA gatescould be used to realize the potential ofQCA for further miniaturization, reduc-tion of power consumption, increase inswitching speed, and increased degreeof integration of very-large-scale inte-grated (VLSI) electronic circuits.

The limitations of conventional VLSIcircuitry, the basic principle of operationof QCA, and the potential advantages ofQCA-based VLSI circuitry were describedin several NASA Tech Briefs articles, namely“Implementing Permutation Matrices byUse of Quantum Dots” (NPO-20801),

Vol. 25, No. 10 (October 2001), page 42;“Compact Interconnection NetworksBased on Quantum Dots” (NPO-20855)Vol. 27, No. 1 (January 2003), page 32;“Bit-Serial Adder Based on QuantumDots” (NPO-20869), Vol. 27, No. 1 (Janu-ary 2003), page 35; and “HybridVLSI/QCA Architecture for ComputingFFTs” (NPO-20923), which follows this ar-ticle. To recapitulate the principle of op-eration (greatly oversimplified because ofthe limitation on space available for thisarticle): A quantum-dot cellular automatacontains four quantum dots positioned ator between the corners of a square cell.The cell contains two extra mobile elec-trons that can tunnel (in the quantum-

mechanical sense) between neighboringdots within the cell. The Coulomb repul-sion between the two electrons tends tomake them occupy antipodal dots in thecell. For an isolated cell, there are two en-ergetically equivalent arrangements (de-noted polarization states) of the extraelectrons. The cell polarization is used toencode binary information. Because thepolarization of a nonisolated cell de-pends on Coulomb-repulsion interac-tions with neighboring cells, universallogic gates and binary wires could be con-structed, in principle, by arraying QCA ofsuitable design in suitable patterns.

Heretofore, researchers have recog-nized two major obstacles to realization ofQCA-based logic gates: One is the need for(and the difficulty of attaining) operationof QCA circuitry at room temperature or,for that matter, at any temperature above afew Kelvins. It has been theorized thatroom-temperature operation could bemade possible by constructing QCA asmolecular-scale devices. However, in ap-proaching the lower limit of miniaturiza-tion at the molecular level, it becomes in-creasingly imperative to overcome thesecond major obstacle, which is the needfor (and the difficulty of attaining) highprecision in the alignments of adjacentQCA in order to ensure the correct inter-actions among the quantum dots.

The fault-tolerant logic gates that wouldbe implemented by blocks of QCA accord-ing to the proposal include majority andinverter (NOT) gates, which are said to beuniversal logic gates because other logicgates (AND, OR, and NOR) can be imple-mented as combinations of majority andinverter gates. The figure depicts exam-ples of (1) a basic QCA majority gate man-ufactured with exact positioning of allQCA, (2) a basic QCA majority gate man-ufactured with a significant position error,(3) a block QCA gate manufactured withexact positioning, and (4) a block QCAgate manufactured with irregularity of po-sitions in the QCA array and errors in thechoice of the edge QCAs used for input.These and other examples were analyzedby computational simulation, using a pro-gram developed at the University of Notre

Basic and Block QCA Majority Gates, with and without misalignments, were analyzed by computa-tional simulation. The results of the analysis showed that, relative to the basic gate, the block gatewould be more tolerant of manufacturing errors.


Dame, that implements a Hartree-Fockmathematical model of the physics of aQCA array. The simulation was performedfor an assumed cell size of 20 nm andinter-cell distance of 14 nm.

The results of the simulation showedthat for a basic QCA majority gate, anoutput error would occur if the errorsin the relative positions of adjacent cellswere to exceed various amounts of theorder of the size of a cell or a significantfraction thereof (the exact amountsbeing different for different cells anddifferent directions of displacement).

In the case of a molecular implementa-tion, this would translate to a require-ment for impractical sub-nanometermanufacturing tolerances. On theother hand, the simulation showed thateven with errors as large as those de-picted for the block majority gate at thebottom of the figure, there would be nooutput error.

This work was done by Amir Firjany,Nikzad Toomarian, and Katayoon Modarresof Caltech for NASA’s Jet Propulsion Lab-oratory. Further information is contained ina TSP (see Page 1).

In accordance with Public Law 96-517, thecontractor has elected to retain title to this in-vention. Inquiries concerning rights for itscommercial use should be addressed to

Intellectual Property groupJPLMail Stop 202-2334800 Oak Grove DrivePasadena, CA 91109(818) 354-2240Refer to NPO-21127, volume and number


A data-processor architecture thatwould incorporate elements of bothconventional very-large-scale inte-grated (VLSI) circuitry and quantum-dot cellular automata (QCA) has beenproposed to enable the highly paralleland systolic computation of fast Fouriertransforms (FFTs). The proposed cir-cuit would complement the QCA-basedcircuits described in several prior NASATech Briefs articles, namely “Implement-ing Permutation Matrices by Use ofQuantum Dots” (NPO-20801), Vol. 25,No. 10 (October 2001), page 42; “Com-pact Interconnection Networks Basedon Quantum Dots” (NPO-20855) Vol.27, No. 1 (January 2003), page 32; and“Bit-Serial Adder Based on QuantumDots” (NPO-20869), Vol. 27, No. 1 (Jan-uary 2003), page 35.

The cited prior articles describedthe limitations of very-large-scale inte-grated (VLSI) circuitry and the majorpotential advantage afforded by QCA.To recapitulate: In a VLSI circuit, sig-nal paths that are required not to in-teract with each other must not crossin the same plane. In contrast, for rea-sons too complex to describe in thelimited space available for this article,suitably designed and operated QCA-based signal paths that are requirednot to interact with each other can nev-ertheless be allowed to cross eachother in the same plane without ad-verse effect. In principle, this charac-teristic could be exploited to designcompact, coplanar, simple (relative toVLSI) QCA-based networks to imple-ment complex, advanced interconnec-tion schemes.

To enable a meaningful descriptionof the proposed FFT-processor archi-tecture, it is necessary to further reca-pitulate the description of a quantum-dot cellular automaton from thefirst-mentioned prior article: A quan-tum-dot cellular automaton containsfour quantum dots positioned at or be-tween the corners of a square cell. Thecell contains two extra mobile electronsthat can tunnel (in the quantum-me-chanical sense) between neighboring

dots within the cell. The Coulomb re-pulsion between the two electronstends to make them occupy antipodaldots in the cell. For an isolated cell,there are two energetically equivalentarrangements (denoted polarizationstates) of the extra electrons. The cellpolarization is used to encode binaryinformation. Because the polarizationof a nonisolated cell depends onCoulomb-repulsion interactions withneighboring cells, universal logic gates

Hybrid VLSI/QCA Architecture for Computing FFTsSimplification is effected through use of QCA circuitry to permute data.NASA’s Jet Propulsion Laboratory, Pasadena, California

Figure 1. QCA Are Assembled Into Binary Wires, and the wires are patterned to implement a perfect-shuffle permutation matrix known as ∏8.

A SYSTOLIC IMPLEMENTATION OF PERMUTATION MATRIX ∏8

a0

a1

a2

a3

a4

a5

a6

a7

a0

Clock 1 Clock 2 Clock 3

QCA Circuit

Schematic Representation

a4

a5

a6

a1

a2

a3

a7

Encoding Binary 1 Encoding Binary 0

CELL POLARIZATION AND BINARY ENCODINGOF INFORMATION IN QCA

CROSSING OF WIRES

Schematic Representation QCA Circuit

1

1

0 0


and binary wires could be constructed,in principle, by arraying QCA of suit-able design in suitable patterns.

Again, for reasons too complex to de-scribe here, in order to ensure accuracyand timeliness of the output of a QCAarray, it is necessary to resort to an adia-batic switching scheme in which theQCA array is divided into subarrays,each controlled by a different phase of amultiphase clock signal. In this scheme,each subarray is given time to performits computation, then its state is frozenby raising its interdot potential barriersand its output is fed as the input to thesuccessor subarray. The successor subar-ray is kept in an unpolarized state so itdoes not influence the calculation ofpreceding subarray. Such a clockingscheme is consistent with pipeline com-putation in the sense that each differentsubarray can perform a different part ofan overall computation. In other words,QCA arrays are inherently suitable forpipeline and, moreover, systolic compu-tations. This sequential or pipeline as-pect of QCA would be utilized in theproposed FFT-processor architecture.

Heretofore, the main obstacle to de-

signing VLSI circuits for systolic andhighly parallel computation of FFTs(and of other fast transforms com-monly used in the processing of imagesand signals) has been the need forcomplex data permutations that cannotbe implemented without crossing ofsignal paths. The proposed hybridVLSI/QCA FFT-processor architecturewould exploit the coplanar-signal-path-crossing capability of QCA to imple-ment the various permutations directlyin patterns of binary wires (that is, lin-ear arrays of quantum dots), as in theexample of Figure 1. The proposed ar-chitecture is based on a reformulationof the FFT by use of a particular matrixfactorization that is suitable for systolicimplementation. The reformulatedFFT is given by

where n is an integer; F2n is a radix-2 FFTfor a 2n-dimensional vector; ∏2n, Si(where i is an integer), and P2n are vari-ous permutation operators or matrices;and the Ki are arithmetic operators.

Figure 2 depicts the proposed archi-tecture. The permutation operatorswould be implemented by QCA mod-ules, while the arithmetic operators Kiwould be implemented by VLSI mod-ules containing simple bit-serial pro-cessing elements. Each processing ele-ment would receive input data from twosources and would produce two outputsby performing simple multiplicationand addition operations. Aside frombeing driven by the same clock (inorder to obtain the necessary globalsynchronization), the processing ele-ments would operate independently ofeach other; because of this feature, theprocessing modules would be amenableto large-scale implementation in com-plementary metal oxide/semiconduc-tor (CMOS) VLSI circuitry. To obtainthe necessary global synchronization,the VLSI and the QCA modules wouldbe driven by the same clock.

This work was done by Amir Fijany,Nikzad Toomarian, Katayoon Modarres,and Matthew Spotnitz of Caltech forNASA’s Jet Propulsion Laboratory. Fur-ther information is contained in a TSP (seepage 1). NPO-20923

F S K S K

S K S K S K S K P

n n

n

n n n n

i i i i

2 2 1 1

1 1 2 2 1 1 2

= ∏ − −

+ +

. . .

. . . ,

Figure 2. A Hybrid of VLSI and QCA Circuit Modules would perform a parallel, systolic computation of an FFT. The particular circuit architecture is basedon a matrix factorization of the FFT.

P2n K1 S1 K2 S2 Ki Si Ki+1 Si +1 Kn –1 Sn –1 Kn Sn ∏2n

Brushlike arrays of carbon nanotubesembedded in microstrip waveguides pro-vide highly efficient (high-Q) mechani-cal resonators that will enable ultra-miniature radio-frequency (RF)integrated circuits. In its basic form, thisinvention is an RF filter based on a car-bon nanotube array embedded in a mi-crostrip (or coplanar) waveguide, asshown in Figure 1. In addition, arrays ofthese nanotube-based RF filters can beused as an RF filter bank.

Applications of this new nanotube arraydevice include a variety of communica-tions and signal-processing technologies.High-Q resonators are essential for stable,low-noise communications, and radar ap-

plications. Mechanical oscillatorscan exhibit orders of magnitudehigher Qs than electronic resonantcircuits, which are limited by resis-tive losses. This has motivated thedevelopment of a variety of me-chanical resonators, including bulkacoustic wave (BAW) resonators,surface acoustic wave (SAW) res-onators, and Si and SiC microma-chined resonators (known as “mi-croelectromechanical systems” orMEMS). There is also a strong pushto extend the resonant frequenciesof these oscillators into the GHzregime of state-of-the-art electron-ics. Unfortunately, the BAW and

Arrays of Carbon Nanotubes as RF Filters in WaveguidesAdvantages would include compactness and high Q.NASA’s Jet Propulsion Laboratory, Pasadena, California

200 nm200 nm

Figure 1. This Array of Carbon Nanotubes, with a diame-ter nonuniformity of <5 percent, was fabricated in aprocess that included the use of a nanopore template (J.Xu et al.).

SAW devices tend to be large and are noteasily integrated into electronic circuits.MEMS structures have been integratedinto circuits, but efforts to extend MEMSresonant frequencies into the GHzregime have been difficult because ofscaling problems with the capacitively-coupled drive and readout. In contrast,the proposed devices would be muchsmaller and hence could be more read-ily incorporated into advanced RF(more specifically, microwave) inte-grated circuits.

During the past few years, techniquesfor fabricating highly-ordered, dense ar-rays of nearly uniform carbon-nanotubecantilevers like so many bristles of a

brush (see Figure 1) have provided theessential basis for this new device. Thebasic principle of operation of such anarray as band-pass filter is excitation of amechanical (acoustic) deformation ofthe nanotubes by an incident RF wave(Figure 2). Coupling between the RFsignal and the nanotubes is provided byCoulomb forces on electric charges inthe nanotubes. The device functions asa narrow-band RF filter because inci-dent waves are reflected from the metal-lic nanotubes, except at the mechanicalresonant frequency of the array. Thehigh-Q mechanical resonance of theuniform nanotube array filters the in-coming RF signal and couples the RF

wave at the resonance frequency intothe output electrode.

The resonance frequency of a nan-otube cantilever depends on its diameterand length. For example, it is estimatedthat the resonance frequency of a carbonnanotube 10 nm in diameter and 100 nmlong would be about 4 GHz. By adjustingthe dimensions of the nanotubes in thearray, it should be possible to select reso-nance frequencies that range from below100 kHz up to tens of GHz.

There have also been attempts to makemechanical resonators using silicon can-tilevers. However, the silicon devices in-vestigated thus far have been limited tooperation at frequencies below 400 MHz,whereas carbon-nanotube devices with Qvalues of the order of 103 at a frequencyof 2 GHz have been demonstrated. More-over, there are experimental data thatsuggest that carbon nanotube resonatorsshould exhibit linear response over alarger dynamic range relative to siliconmechanical resonators.

This work was done by Daniel Hoppe,Brian Hunt, Michael Hoenk, and FlavioNoca of Caltech for NASA’s Jet PropulsionLaboratory and by Jimmy Xu of Brown Uni-versity. Further information is contained in aTSP (see Page 1).





Carbon Nanotubes

Incoming Broad-BandRF Wave

ConductingMicrostrip

Dielectric

GroundPlane

BaseElectrode

Outgoing Narrow-BandRF Wave

Figure 2. A Brushlike Array of Carbon Nanotubes embedded in a microstrip waveguide would act asa band-pass filter.

Electromechanical resonators of a pro-posed type would comprise single carbonnanotubes suspended between electrodes(see Figure 1). Depending on the nan-otube length, diameter, and tension, thesedevices will resonate at frequencies in arange from megahertz through gigahertz.Like the carbon-nanotube resonators de-scribed in the preceding article, these de-vices will exhibit high quality factors (Q

values), will be compatible with integra-tion with electronic circuits, and, unlikesimilar devices made from silicone and sil-icone carbide, will have tunable resonantfrequencies as high as several GHz.

An efficient electromechanical trans-duction method for the carbon nanotuberesonators is provided by the previouslyobserved variation of carbon nanotubelength with charge injection. It was found

that injection of electrons or holes, re-spectively, lengthens or shortens carbonnanotubes, by amounts of the order of apercent at bias levels of a few volts. Thecharge-dependent length change also en-ables a simple and direct means of tuningthe resonant frequency by varying the DCbias and hence the tension along thetube, much like tuning a guitar string.

In its basic form, the invention is a tun-

Carbon Nanotubes as Resonators for RF Spectrum AnalyzersCompact, high-speed, high-Q spectrum analyzers could be integrated with other circuits.NASA’s Jet Propulsion Laboratory, Pasadena, California

able high-Q resonator based on a sus-pended carbon nanotube bridge with at-tached electrodes (see Figure 1). An ap-plied DC bias controls the tension and

thus the frequency of resonance. If onewere to superimpose a radio-frequency(RF) bias on the DC bias, then the result-ing rapid variation in tension or length

would set the tube into vibration. If, on theother hand, the carbon nanotube were tobe set into vibration by interaction be-tween an incident RF electric field andelectric charges in the nanotube, then thevibration would give rise to an RF signaloutput that is proportional to the RF am-plitude at the resonance frequency.

Because the transduction mechanism isextremely sensitive and the active volume isonly a few nanometers in diameter, this de-vice is not well suited for use as a microwavepower device. Instead, this carbon nan-otube mechanical resonator would be use-ful primarily as part of a highly precise, sen-sitive, frequency-selective detector. An arrayof such devices featuring nanotubes of dif-ferent lengths (and thus different frequen-cies) could be made to operate as a high-speed spectrum analyzer (see Figure 2).

This work was done by Brian Hunt, FlavioNoca, and Michael Hoenk of Caltech forNASA’s Jet Propulsion Laboratory. Fur-ther information is contained in a TSP (seepage 1).





ElectrodeCarbon

Nanotube

CarbonNanotube

Input Electrode

Output Electrodes

Figure 1. A Carbon Nanotube Suspended Between Electrodes could be stretched taut so that it wouldresonate in the same manner as that of a string on a musical instrument. It could serve as a tunable,high-Q resonator for a signal processor or a sensor.

Figure 2. An Array of Devices like that of Figure 1, containing nanotubes of different lengths wouldbe combined with an input electrode to construct an electromagnetic-spectrum analyzer that wouldfunction somewhat like a cochlea.


Software

Software for Viewing LandsatMosaic Images

A Windows-based computer programhas been written to enable novice users(especially educators and students) to viewimages of large areas of the Earth (e.g., thecontinental United States) generated fromimage data acquired in the Landsat obser-vations performed circa the year 1990.The large-area images are constructed asmosaics from the original Landsat images,which were acquired in several wavelengthbands and each of which spans an area (ineffect, one tile of a mosaic) of ≈5° in lati-tude by ≈6° in longitude. Whereas theoriginal Landsat data are registered on auniversal transverse Mercator (UTM) grid,the program converts the UTM coordi-nates of a mouse pointer in the image tolatitude and longitude, which are continu-ously updated and displayed as the pointeris moved. The mosaic image currently ondisplay can be exported as a Windows bit-map file. Other images (e.g., of stateboundaries or interstate highways) can beoverlaid on Landsat mosaics. The programinteracts with the user via standard toolbar,keyboard, and mouse user interfaces. Theprogram is supplied on a compact diskalong with tutorial and educational infor-mation.

This program was written by Zack Watts,Catharine L. Farve, and Craig Harvey ofPixSell, Inc., for Stennis Space Center.

In accordance with Public Law 96-517, thecontractor has elected to retain title to this in-vention. Inquiries concerning rights for its com-mercial use should be addressed to

PixSell, Inc.Building 2105Stennis Space Center, MS 39529Refer to SSC-00148, volume and number


Updated Integrated MissionProgram

Integrated Mission Program (IMP) is acomputer program for simulating space-craft missions around the Earth, Moon,Mars, and/or other large bodies. IMPsolves the differential equations of motionby use of a Runge-Kutta numerical-inte-gration algorithm. Users control missionsthrough selection from a large menu ofevents and maneuvers. Mission profiles,

time lines, propellant requirements, feasi-bility analyses, and perturbation analysescan be computed quickly and accurately. Aprior version of IMP, written in FORTRAN77, was reported in “Program SimulatesSpacecraft Missions” (MFS-28606), NASATech Briefs, Vol. 17, No. 4 (April 1993), page60. The present version, written in double-precision Lahey™ FORTRAN 90, incorpo-rates a number of improvements over theprior version. Some of the improvementsmodernize the code to take advantage oftoday’s greater central-processing-unitspeeds. Other improvements render thecode more modular; provide additionalinput, output, and debugging capabilities;and add to the variety of maneuvers,events, and means of propulsion that canbe simulated. The IMP user manuals (ofwhich there are now ten, each addressinga different aspect of the code and its use)have been updated accordingly.

This program was written by Vincent A.Dauro, Sr., of Alpha Technology, Inc., for Mar-shall Space Flight Center. Further informa-tion is contained in a TSP (see page 1).MFS-31695

Software for Sharing andManagement of Information

DIAMS is a set of computer programsthat implements a system of collaborativeagents that serve multiple, geographicallydistributed users communicating via theInternet. DIAMS provides a user interfaceas a Java applet that runs on each user’scomputer and that works within the con-text of the user’s Internet-browser soft-ware. DIAMS helps all its users to manage,gain access to, share, and exchange infor-mation in databases that they maintain ontheir computers. One of the DIAMSagents is a personal agent that helps itsowner find information most relevant tocurrent needs. It provides software toolsand utilities for users to manage their in-formation repositories with dynamic orga-nization and virtual views. Capabilities forgenerating flexible hierarchical displaysare integrated with capabilities for in-dexed-query searching to support effectiveaccess to information. Automatic indexingmethods are employed to support users’queries and communication betweenagents. The catalog of a repository is keptin object-oriented storage to facilitate shar-ing of information. Collaboration between

users is aided by matchmaker agents andby automated exchange of information.The matchmaker agents are designed toestablish connections between users whohave similar interests and expertise.

This program was written by James R. Chenof Ames Research Center and Shawn R.Wolfe and Stephen D. Wragg of Caelum Re-search. Further information is contained in aTSP (see page 1).ARC-14654

Update on Integrated Optical Design Analyzer

Updated information on the Inte-grated Optical Design Analyzer (IODA)computer program has become available.IODA was described in “Software for Mul-tidisciplinary Concurrent Optical Design”(MFS-31452), NASA Tech Briefs, Vol. 25,No. 10 (October 2001), page 8a. To reca-pitulate: IODA facilitates multidiscipli-nary concurrent engineering of highlyprecise optical instruments. The architec-ture of IODA was developed by reviewingdesign processes and software in an effortto automate design procedures. IODAsignificantly reduces design iterationcycle time and eliminates many potentialsources of error. IODA integrates themodeling efforts of a team of experts indifferent disciplines (e.g., optics, struc-tural analysis, and heat transfer) workingat different locations and provides seam-less fusion of data among thermal, struc-tural, and optical models used to designan instrument. IODA is compatible withdata files generated by the NASTRAN™structural-analysis program and the CodeV® optical-analysis program, and can beused to couple analyses performed bythese two programs. IODA supports mul-tiple-load-case analysis for quickly accom-plishing trade studies. IODA can alsomodel the transient response of an in-strument under the influence of dynamicloads and disturbances.

This program was written by James D. Moore,Jr., and Ed Troy of SRS Technologies for Mar-shall Space Flight Center. Further informa-tion is contained in a TSP (see page 1).MFS-31809


Materials

A method of fabricating both curvedand flat thin polymer membranes of op-tical quality has been developed. Themethod was originally intended to en-able the fabrication of lightweight mem-brane imaging and interferometric op-tics, possibly with apertures multiplemeters wide, for use in scientific instru-ments that would operate in outer space.The method may also be applicable tothe fabrication of lightweight membraneoptics for terrestrial use.

The method involves flow-casting of asoluble polymer with mechanical and

environmental controls that providenearly ideal conditions for the formationof a membrane. The preferred environ-mental conditions and other details ofthe process depend on the choice ofpolymer and substrate material and onthe shape and size of the membrane tobe cast. Once the polymer has dried to amembrane, it is cured with convectiveheating, then released.

Membranes with root-mean-squaresurface roughnesses of <10.5 Å can beproduced routinely by this method. Vari-ations in the thicknesses of the mem-

branes have ranged from 1/3 wave-length down to as little as 1/20 wave-length (at a wavelength of 633 nm).Membranes fabricated thus far have haddiameters up to 0.5 m, and there ap-pears to be no major obstacle to scalingup to multiple-meter diameters.

This work was done by James Moore andBrian Patrick for Marshall Space FlightCenter. Further information is contained ina TSP (see Page 1).

MFS-31718

Optical-Quality Thin Polymer MembranesSurface roughnesses and thickness variations are small enough for demanding scientificapplications.Marshall Space Flight Center, Alabama


Mechanics

Rollable Thin Shell Composite-Material Paraboloidal MirrorsThese lightweight focusing mirrors can be stored in fairly narrow cylinders.NASA’s Jet Propulsion Laboratory, Pasadena, California

An experiment and calculation havedemonstrated the feasibility of a tech-nique of compact storage of paraboloidalmirrors made of thin composite-material(multiple layers of carbon fiber mats in apolymeric matrix) shells coated withmetal for reflectivity. Such mirrors areunder consideration as simple, light-weight alternatives to the heavier, morecomplex mirrors now used in space tele-scopes. They could also be used on Earthin applications in which gravitational sagof the thin shells can be tolerated.

The present technique is essentially thesame as that used to store large maps,posters, tapestries, and similar objects: Onesimply rolls up the mirror to a radius smallenough to enable the insertion of the mir-ror in a protective cylindrical case. Pro-vided that the stress associated with rollingthe mirror is not so large as to introduce anappreciable amount of hysteresis, the mir-ror can be expected to spring back to itsoriginal shape, with sufficient precision toperform its intended optical function,when unrolled from storage.

A simple calculation yields a qualitativeindication of the level of stress in, and thelikelihood of permanent deformation of,a rolled mirror. The calculation in ques-tion is an estimate of the stress in a rolledflat sheet of the same composite materialand thickness as those of the mirror shell.The compressive or tensile stress (S) inthe radially innermost or radially outer-most surface layer, respectively, is given byS = Et/2r, where E is the modulus of elas-ticity of the composite-material shell orflat sheet, t is the thickness of the shell orflat sheet, and r is the radius of curvature

to which the shell or sheet is rolled. For atypical mirror diameter (D = 2 m) andshell thickness (t = 1 mm) rolled to a ra-dius such that diametrically oppositepoints on the edge of the mirror justcome into contact (r = D/2π), this equa-tion yields S ≈ 0.016E. This is a relativelysmall amount of stress and, as such,would not be expected to cause an ap-preciable permanent deformation.

The figure depicts stages of a demon-stration in which a composite-material

mirror of D = 90 cm, t = 1 mm, and a focalratio (f number) of 1 was manually rolledas described above. Visual inspectionafter unrolling revealed no hysteresis.Further optical testing of the unrolledmirror was underway at the time of re-porting the information for this article.

This work was done by Aden Meinel, MarjorieMeinel, and Robert Romeo of Caltech for NASA’sJet Propulsion Laboratory. Further informa-tion is contained in a TSP (see page 1).NPO-20987

A Thin-Shell Composite-Material Mirror was rolled, then unrolled without introducing visible perma-nent distortion of its optical surface.

Mirror Before Rolling

Mirror After Rolling End and Side Views of Rolled-Up Mirror

Folded Resonant Horns for Power Ultrasonic ApplicationsUltrasonic actuators can be made shorter.NASA’s Jet Propulsion Laboratory, Pasadena, California

Folded horns have been conceivedas alternatives to straight horns used asresonators and strain amplifiers inpower ultrasonic systems. Such systemsare used for cleaning, welding, solder-

ing, cutting, and drilling in a variety ofindustries. In addition, several previ-ous NASA Tech Briefs articles have de-scribed instrumented drilling, coring,and burrowing machines that utilize

combinations of sonic and ultrasonicvibrational actuation. The main advan-tage of a folded horn, relative to astraight horn of the same resonancefrequency, is that the folded horn can


be made shorter (that is, its greatestlinear dimension measured from theoutside can be made smaller). Alterna-tively, for a given length, the resonancefrequency can be reduced. Hence, thefolded-horn concept affords an addi-tional degree of design freedom for re-ducing the length of an ultrasonicpower system that includes a horn.

Figure 1 depicts an ultrasonic actua-tor that includes a straight steppedhorn, one that includes an inverted

straight stepped horn of approximatelythe same resonance frequency, and onethat includes a folded stepped horn ofapproximately the same resonance fre-quency. The main role of the straightstepped horn is to amplify longitudinalstrain at its outermost end. In thefolded version, one can exploit bendingstrain in addition to longitudinal strain,and by adjusting the thickness of thefolds, one can increase or decrease thecontributions of bending displacements

to the overall displacement at the tip. Inthis case, the folded-horn concept notonly yields a shorter horn, but by en-abling utilization of bending displace-ments, it also affords an additional de-gree of design freedom. Figure 2 showsan experimental folded-horn actuatorof 16-kHz resonance frequency along-side a straight-horn actuator of 20-kHzresonance frequency.

This work was done by Stewart Sherrit,Stephen Askins, Michael Gradziel, XiaoqiBao, Zensheu Chang, Benjamin Dolgin, andYoseph Bar-Cohen of Caltech and Tom Peter-son of Cybersonics Inc. for NASA’s JetPropulsion Laboratory. Further informa-tion is contained in a TSP (see page 1).NPO-30489

2 cm

16-kHz ACTUATOR 20-kHz ACTUATOR

Figure 1. Three Similar Power Ultrasonic Actuators are depicted partly in cross sections to illustrate aprogression of designs from a straight stepped horn to a folded inverted stepped horn.

Figure 2. The Overall Length of the 16-kHz HornIs Shorter than the 20-kHz horn by virtue ofbeing folded. The distance the acoustic wavetravels has been designed to be the same. Thelower frequency in the folded horn is due to re-duced clamping and bending at the folds.

Stepped Horn

Inverted Stepped Horn

Folded InvertedStepped Horn

Stacked Piezoelectric Disks

Bolt

Nut


Machinery/Automation

The torque-limited touchdown bearingsystem (TLTBS) is a backup mechanical-bearing system for a high-speed rotary ma-chine in which the rotor shaft is supportedby magnetic bearings in steady-state nor-mal operation. The TLTBS provides ball-bearing support to augment or supplantthe magnetic bearings during startup,shutdown, or failure of the magnetic bear-ings. The TLTBS also provides support inthe presence of conditions (in particular,rotational acceleration) that make it diffi-cult or impossible to control the magneticbearings or in which the magnetic bear-ings are not strong enough (e.g., when theside load against the rotor exceeds theavailable lateral magnetic force).

The TLTBS includes two similar or iden-tical subsystems, each located at one end ofthe rotor shaft (see figure). Mounted in-side each female cone is a specially de-signed high-speed bearing with a built-in

lubrication system. Mounted on the statoris (1) a non-rotating drive cone capable ofmating with the drive cone on the rotor and(2) an electromagnetically actuated insertermechanism that moves the stator drive coneaxially between two extreme positions asdescribed in more detail in the next para-graph. The inserter mechanism containstwo electromagnets: one that withdraws thespring-loaded plunger, the second to with-draw the latch when insertion is required.Electric power is applied to the insertermechanism only during the fraction of asecond needed for the axial motion: nopower is needed to keep the stator drivecone latched at either extreme position.

In one extreme axial position, denotedthe inserted position, the stator drivecone is in contact with the outer-bearing-race drive cone under spring load. In thisposition, the ball-bearing assembly car-ries the full bearing load.

The other extreme axial position (theone shown in the figure) is denoted theretracted position. In this position, the sta-tor drive cone is withdrawn from the rotordrive cone to an axial clearance of 0.010in. (≈0.25 mm). During normal operationin the retracted position, the shaft is fullysupported by the magnetic bearing. Theclearance between drive cones is smallenough that if the magnetic bearing failsor if an excessive side load occurs, the ball-bearing assembly can provide full support,preventing damage to the magnetic sus-pension and making it possible to con-tinue (at least temporarily) operation ofthe machine.

Because the stator drive cone does notrotate and the rotor drive cone rotates athigh speed during normal operation of themagnetic bearing, it is necessary to accom-modate, and prevent frictional damage by,the slip that occurs between the two drive

The Portion of the TLTBS at One End of a Rotor Shaft includes a ball-bearing assembly and drive cones that can be either mated or separated by a smallclearance as shown here.

Magnetic-BearingRotor

Inner Race

LubricationReservoir

Drive Cones

Spring

Anti-Friction Coating

Outer Race

LatchingPlunger

Electromagnet forAxial Motion

Pin To Prevent Rotation ofStator Drive Cone

Spring

Magnetic-BearingStator

Rotor Wheel

Shaft

Touchdown Ball-Bearing System for Magnetic BearingsIn the event of a touchdown, ball bearings provide full support.John H. Glenn Research Center, Cleveland, Ohio


cones at the time of initial contact. For thispurpose, the mating surfaces of the drivecones are coated with dry lubricant films. Inaddition, the ball-bearing assembly containsa reservoir designed to dispense lubricantto support an elastohydrodynamic film forthe specified lifetime of the system.

This work was done by Edward P. Kings-bury, Robert Price, Erik Gelotte, and Herbert B.Singer of The Bearing Consultants, LLP forGlenn Research Center. Further informa-tion is contained in a TSP (see page 1).

Inquiries concerning rights for the com-mercial use of this invention should be ad-

dressed to NASA Glenn Research Center,Commercial Technology Office, Attn: SteveFedor, Mail Stop 4–8, 21000 BrookparkRoad, Cleveland, Ohio 44135. Refer toLEW-17282.

Flux-Based Deadbeat Control of Induction-Motor TorqueGraphical constructions provide insight into solutions of control problems.John H. Glenn Research Center, Cleveland, Ohio

An improved method and prior meth-ods of deadbeat direct torque control in-volve the use of pulse-width modulation(PWM) of applied voltages. The priormethods are based on the use of statorflux and stator current as state variables,leading to mathematical solutions of con-trol equations in forms that do not lendthemselves to clear visualization of solu-tion spaces. In contrast, the use of rotorand stator fluxes as the state variables inthe present improved method lends itselfto graphical representations that aid inunderstanding possible solutions undervarious operating conditions. In addi-tion, the present improved method in-corporates the superposition of high-fre-quency carrier signals for use in amotor-self-sensing technique for estimat-ing the rotor shaft angle at any speed (in-cluding low or even zero speed) withoutneed for additional shaft-angle-measur-ing sensors.

Prerequisite to a description of themethod is a description of the conceptof dq variables and dq coordinates. The“d” and “q” signify “direct” and “quadra-ture,” respectively. The dq coordinateslie along two orthogonal axes attachedto the stator. The dq coordinates and thedq variables (which can be, for example,phase voltages and fluxes projectedonto the dq coordinates) are commonlyused in the design and analysis of in-duction motors.

The derivation of the method beginswith the observation that a desired changein the torque of an induction motor can berepresented as a straight line with units offlux on both axes of the d-q plane. Theequation that describes the straight linecan be derived from the discrete form ofthe induction-motor equations by use ofthe stator and rotor fluxes as state variables.The desired change in stator flux can berepresented as a circle on the d-q plane.The voltage needed to achieve the desiredchange in torque and change in stator flux

in one time step can be found from an in-tersection of the torque line and stator-fluxcircle (see figure).

Going beyond the aforementionedgraphical representations, the maximum-current operating limits can be repre-sented as a set of two lines parallel to thetorque line. The maximum-voltage operat-ing limits can be represented as a hexagonon the d-q plane. The maximum-voltagehexagon can be divided into four regionscorresponding to an increase or decreasein torque and an increase or decrease influx. As a result, the effect of a particularvoltage vector (increase or decrease intorque or flux) can be clearly seen.

A control algorithm based on thegraphical constructions and the underly-ing equations has been developed. Thealgorithm causes the synthesis of theneeded excitation voltages by use ofspace vector modulation techniques tocalculate and command inverter duty cy-

cles. In the implementation of the algo-rithm without shaft-angle-measuringsensors, the high-frequency voltageneeded for sensorless operation is addedto the fundamental voltage and used asinput for the PWM calculations. ThePWM portion of the control algorithmthen determines the duty cycles to gen-erate both voltages at once. The signalfrom the resulting high-frequency cur-rent is processed to estimate the angularposition and speed of the rotor.

This work was done by Barbara H. Kennyof Glenn Research Center and Robert D.Lorenz of the University of Wisconsin. Fur-ther information is contained in a TSP (seepage 1).

Inquiries concerning rights for the commer-cial use of this invention should be addressedto NASA Glenn Research Center, CommercialTechnology Office, Attn: Steve Fedor, MailStop 4–8, 21000 Brookpark Road, Cleve-land, Ohio 44135. Refer to LEW-17329.

–0.10

–0.08

–0.06

–0.04

–0.02

0

0.02

0.04

–0.05 0

q-Axis Flux (Example Values in Webers)

d-A

xis

Flu

x (E

xam

ple

Val

ues

in W

eber

s)

0.05

Voltage Vector

A Straight Line and a Circle in the d-q Plane represent the desired change in torque and the desiredchange in stator flux, respectively. The voltage vector needed to achieve the desired change in torque andchange in stator flux in one time step can be found from one of the intersections of line and the circle.


A method of manufacturing regular ar-rays of precisely sized, shaped, positioned,and oriented carbon nanotubes has beenproposed. Arrays of carbon nanotubescould prove useful in such diverse appli-cations as communications (especially forfiltering of signals), biotechnology (for se-quencing of DNA and separation ofchemicals), and micro- and nanoelectron-ics (as field emitters and as signal trans-ducers and processors). The method is ex-pected to be suitable for implementationin standard semiconductor-device fabrica-tion facilities.

As in previously reported methods,carbon nanotubes would be formed inthe proposed method by decompositionof carbon-containing gases overnanometer-sized catalytic metal particlesthat had been deposited on suitable sub-strates. Unlike in previously reportedmethods, the catalytic metal particleswould not be so randomly and denselydistributed as to give rise to thick, irreg-ular mats of nanotubes with a variety oflengths, diameters, and orientations. In-stead, in order to obtain regular arrays ofspaced-apart carbon nanotubes as nearlyidentical as possible, the catalytic metalparticles would be formed in predeter-mined regular patterns with precisespacings. The regularity of the arrayswould be ensured by the use of nanos-tructured templates made of blockcopolymers.

A block copolymer consists of two ormore sections, or “blocks,” each of whichconsists of a controlled number ofmonomers of a given type. Some combi-nations of monomers (for example,styrene and methylmethacrylate or styreneand butadiene) yield block copolymermolecules that, under appropriate condi-tions (for example, when heated abovetheir glass-transition temperatures for 10to 20 hours) assemble themselves into re-peating structures with unit-cell dimen-sions that typically range between 5 and100 nm. In other words, a block copoly-mer can be made to acquire a regularstructure on a length scale substantiallylarger than the individual monomer units

Manufacturing

Block Copolymers as Templates for Arrays of CarbonNanotubesThe spontaneous formation of nanostructures in block copolymers would be exploited.NASA’s Jet Propulsion Laboratory, Pasadena, California

A Template for Deposition of catalytic metal particles would be formed in a PS/PMMA block copoly-mer, then carbon nanotubes would be grown on the particles by decomposition of carbon-containinggas molecules.

PSPS

PM

MA

PM

MA

PM

MA

PM

MA

PM

MA

PM

MA

PSPS PSPSPS PSPS

Substrate

CarbonNanotubes

Catalytic Metal Particles

PS


yet well below a macroscopic scale. For ex-ample, when heated in an electric field, athin film of molecules of a block copoly-mer consisting of sections of polystyrene(PS) connected to sections of poly(methyl-methacrylate) [PMMA] undergoes ananoscale phase separation, formingaligned, regularly-spaced cylinders ofPMMA spaced apart in a matrix of PS. Thesize and separation of the cylinders de-pends on the sizes of the PS and PMMAblocks in the molecules.

Proposed techniques for utilizing suchnanostructured block copolymers astemplates are generally oriented towardexploiting the differences betweenchemical and/or physical properties ofthe different materials in the adjacentnanoscale regions. In particular, onecould utilize differences in reactivities ofthe blocks with respect to one or morechemical(s) in order to selectively re-move the blocks of one type, without re-moving the adjacent blocks of the other

type, in order to create voids into whichcatalytic metals could be deposited.

In a typical application (see figure),one would begin by coating a substratewith a film of PS/PMMA diblock copoly-mer. (A typical substrate material wouldbe silicon with a surface layer of siliconoxide.) The block copolymer film wouldbe treated to form the desired repeatingnanostructure. The nanostructuredblock copolymer would be exposed to ul-traviolet light or to high-energy elec-trons, either of which would degrade thePMMA more than it would degrade thePS. The workpiece would be washed inacetic acid to remove the degradedPMMA, yielding a substrate covered witha thin film of polystyrene containing aregular array of nanometer-sized holes,into which catalytic metal could be de-posited either electrochemically or fromvapor. The polystyrene would be re-moved from around the deposited metalparticles by oxidation, solvation, or etch-

ing with reactive ions. Remaining on thesubstrate would be a regular array of cat-alytic metal dots, on which a regulararray of carbon nanotubes could then begrown.

This work was done by MichaelBronikowski and Brian Hunt of Caltech forNASA’s Jet Propulsion Laboratory. Fur-ther information is contained in a TSP (seepage 1).





Physical Sciences

An improved design has been pro-posed for a cryostat of a type that main-tains a desired low temperature mainlythrough boiloff of a liquid cryogen (e.g.,liquid nitrogen) at atmospheric pres-sure. (A cryostat that maintains a lowtemperature mainly through boiloff of acryogen at atmospheric pressure is saidto be of the pour/fill Dewar-flask typebecause its main component is a Dewarflask, the top of which is kept open tothe atmosphere so that the liquid cryo-gen can boil at atmospheric pressureand cryogenic liquid can be added bysimply pouring it in.) The major distin-guishing feature of the proposed designis control of temperature and coolingrate through control of the flow of cryo-gen vapor from a heat exchanger. At acost of a modest increase in complexity,a cryostat according to the proposalwould retain most of the compactness ofprior, simpler pour/fill Dewar-flaskcryostats, but would utilize cryogen moreefficiently (intervals between cryogen re-fills could be longer).

In a typical prior variable-temperaturecryostat of the pour/fill Dewar-flask type,a specimen (typically, an infrared pho-todetector) to be tested at a specifiedlow temperature is mounted, along witha small electric heater, on a variable-tem-perature stage on one side of a thermal-resistance block. The other side of thethermal-resistance block is in contactwith the cold inner wall of the cryogenreservoir (see figure). If it is desired totest the specimen at a temperature abovethe boiling temperature of the cryogen,then a proportional amount of power issupplied to the electric heater. Ofcourse, the heat that leaks through thethermal-resistance block into the cryo-gen reservoir causes the cryogen to boiloff faster than it otherwise would.

In the proposed cryostat, the variable-temperature stage holding the speci-men and electric heater would bemounted on one surface of a heat ex-

changer that would be held away fromthe cryogen reservoir by high-thermal-resistance standoffs. A tube would sup-ply liquid cryogen from the bottom of

the cryogen reservoir, upward into theheat exchanger, by gravity feed. Fromthe top of the heat-exchanger, boiled-offcryogen vapor would travel through a

tube and a valve, returningto the top of the reservoir.By using the valve to throttlethe flow of vapor from theheat exchanger, one wouldcontrol the flow of liquid tothe heat exchanger, therebycontrolling the rate of cool-ing and the temperature ofthe specimen. In addition,as before, one could raisethe temperature of the spec-imen by use of the electricheater. To the extent towhich one could maintainthe desired temperature byflow control rather thanelectric heating, one couldreduce the consumption ofboth liquid cryogen andelectrical energy.

This work was done byThomas Cunningham of Cal-tech for NASA’s Jet Propul-sion Laboratory. Further in-formation is contained in aTSP (see page 1).NPO-21186

Throttling Cryogen Boiloff To Control Cryostat TemperatureConsumption of liquid cryogen and electrical energy could be reduced.NASA’s Jet Propulsion Laboratory, Pasadena, California

Liquid Cryogen

Liquid-Cryogen Supply Tube

HeatExchanger

ElectricHeater

CryogenReservoir

High-Thermal-ResistanceStandoffs

Variable-Temperature

Stage

Specimen

Electrical Feedthroughs

Valve

PROPOSED DESIGN

Electrical Feedthroughs

VacuumSpace

Liquid CryogenCryogenReservoir

Thermal-Resistance Block

Variable-Temperature

Stage

Specimen

ElectricHeater

PRIOR DESIGN

These Variable-Temperature Cryostats of the pour/fill-Dewar-flask type differ in the nature of the thermal resistance be-tween the cryogen reservoir and the variable-temperaturestage. The fixed thermal resistance of the prior design would bereplaced, in the proposed design, by a variable thermal resis-tance implemented by use of a heat exchanger with a flow-con-trol valve.


Information Sciences

United Space Alliance (USA) devel-oped and used a new software develop-ment method to meet technical, sched-ule, and budget challenges faced duringthe development and delivery of the newShuttle Telemetry Ground Station atKennedy Space Center. This method,called Collaborative Software Develop-ment, enabled KSC to effectively leverageindustrial software and build additionalcapabilities to meet shuttle system and op-erational requirements. Application ofthis method resulted in reduced time tomarket, reduced development cost, im-proved product quality, and improved

programmer competence while develop-ing technologies of benefit to a small com-pany in California (AP Labs Inc.). Manymodifications were made to the baselinesoftware product (VMEwindow), whichimproved its quality and functionality. Inaddition, six new software capabilitieswere developed, which are the subject ofthis article and add useful functionality tothe VMEwindow environment. These newsoftware programs are written in C or VX-Works and are used in conjunction withother ground station software packages,such as VMEwindow, Matlab, Dataviews,and PVWave.

The Space Shuttle Telemetry GroundStation receives frequency-modulation(FM) and pulse-code-modulated (PCM)signals from the shuttle and support equip-ment. The hardware architecture (see fig-ure) includes Sun workstations connectedto multiple PCM- and FM-processing Ver-saModule Eurocard (VME) chassis. A re-flective memory network transports rawdata from PCM Processors (PCMPs) to theprogrammable digital-to-analog (D/A)converters, strip chart recorders, and analy-sis and controller workstations.

The first new program providesVMEwindow access to the reflective

Collaborative Software Development Approach Used To Deliverthe New Shuttle Telemetry Ground StationThis software affords enhanced capabilities for utilizing telemetric data.John F. Kennedy Space Center, Florida

GPIBConverter

GPIBConverter

GPIBConverter

FM Switch

FMP

D/AConverters

Strip-Chart andThermal Array

Recorders

Station-Controller

Workstations

User AnalysisWorkstations

Data-BaseHost

Computer

Printer ServerComputer

Color Plotters Line Printers Laser Printers

User PersonalComputer

PCMP

Data and Reflective Memory Network

Set Management and Control Network

Time Switch PCM Switch

Signal Patch Equipment

Input Signals

Recorders andPlayers

Router

The Upgraded RPS includes, notably, a reflective memory network that serves as the primary means of transport of raw data.


memory via a Sun Sbus interface. Thisprogram makes it possible to acquireand display data from multiple real-timetelemetry processors, with deterministic,minimal latency, at any of the ground-station Sun workstations.

Another program prompts the user toenter a mission number, parameters andmeasurement names for a specific PCMtelemetry stream. The program thenqueries an Oracle database and createsa flat file with the appropriate informa-tion for setup of VME chassis. This file isthen automatically imported to thesetup configuration in VMEwindow.Prior to this development, the user wasrequired to enter all stream parametersvia a keyboard. This was an extremelytime-consuming and laborious process,which was very prone to error. The pro-gram also retrieves and loads data to au-tomatically generate trigger parameters,derived parameters, and mathematicalformulas needed to process the re-trieved data.

The third program affords capabilitiesfor setting up and controlling a general-purpose interface bus (GPIB) circuitboard to perform special FM-signal-pro-cessing functions. These functions in-clude snapshots that provide average,minimum, and maximum values of 100

samples of data, and calibrations thatprovide an average of 5 samples at eachof the 0-, 25-, 50-, 75-, and 100-percentdata levels. The setup enables individualcontrol of the discriminator and theswitch, along with automated control ofsnapshots and calibrations. The setupfunctions of this program are integratedinto the VMEwindow telemetry-controlsoftware and are accessible via an icon ina VMEwindow display. The control func-tions of this program are exerted via thedriver software of the GPIB circuit board.

The fourth program provides for thesetup and control of a digital-to-analogconverter. As in the case of the thirdprogram, the setup functions of thisprogram are integrated into theVMEwindow telemetry-control softwareand are accessible via an icon in aVMEwindow display. Control is pro-vided by a driver subprogram that in-terprets the setup and controls theboard accordingly to make it generatethe required output.

The fifth program provides the abilityto reconstitute major frames of data fromsingle minor frame PCM data outputfrom the PCMP decom. Shipping minorframes of data was required to provideminimum data latency and loss to the stripchart recorders in the event of a loss of the

telemetry signal. Major frames of minorframe data are fed to VMEwindow engi-neering conversion icons in order to takeadvantage of the displays and processingcapabilities of VMEwindow.

The final software innovation “thus far”provides the capability to record selectedparameters in real time and output thisdata to a formatted text file. This file canbe viewed displaying near-time history ofparameter values with associated timetags. The data can be displayed in deci-mal, hexadecimal, octal, binary, and engi-neering converted formats and either alldata or change data can be stored.

The baseline ground station develop-ment effort spawned the first four ofthese innovations discussed. The last twodiscussed were developed under a con-tinuing collaborative relationship be-tween USA and AP Labs Inc., designedto improve shuttle processing and fostercontinued technological innovation.

This work was done by Randy L. Kirby,David Mann, Stephen G. Prenger, WayneCraig, Andrew Greenwood, Jonathan Mor-sics, and Charles H. Fricker of United SpaceAlliance and Son Quach and Paul Lechese ofAP Labs for Kennedy Space Center. Fur-ther information is contained in a TSP (seepage 1).KSC-12100/01/02/03/71/81


Turbulence in SupercriticalO2/H2 and C7H16/N2 Mix-ing Layers

This report presents a study of numer-ical simulations of mixing layers develop-ing between opposing flows of paired flu-ids under supercritical conditions, thepurpose of the study being to elucidatechemical-species-specific aspects of tur-bulence. The simulations were per-formed for two different fluid pairs —O2/H2 and C7H16/N2 — at similar re-duced initial pressures (reduced pres-sure is defined as pressure ÷ critical pres-sure). Thermodynamically, O2/H2behaves more nearly like an ideal mix-ture and has greater solubility, relative toC7H16/N2, which departs strongly fromideality. Because of a specified smallerinitial density stratification, theC7H16/N2 layers exhibited greater levelsof growth, global molecular mixing, andturbulence. However, smaller density gra-dients at the transitional state for theO2/H2 system were interpreted as indi-cating that locally, this system exhibits en-hanced mixing as a consequence of itsgreater solubility and closer approach toideality. These thermodynamic featureswere shown to affect entropy dissipation,which was found to be larger for O2/H2and concentrated in high-density-gradi-ent-magnitude regions that are distor-tions of the initial density-stratificationboundary. In C7H16/N2, the regions oflargest dissipation were found to lie inhigh-density-gradient-magnitude regionsthat result from mixing of the two fluids.

This work was done by Josette Bellan, Ken-neth Harstad, and Nora Okong’o of Caltechfor NASA’s Jet Propulsion Laboratory.Further information is contained in a TSP(see page 1).NPO-30561

Time-Resolved Measure-ments in Optoelectronic Microbioanalysis

A report presents discussion of time-re-solved measurements in optoelectronicmicrobioanalysis. Proposed microbioana-lytical “laboratory-on-a-chip” devices fordetection of microbes and toxic chemicalswould include optoelectronic sensors andassociated electronic circuits that wouldlook for fluorescence or phosphorescencesignatures of multiple hazardous biomole-cules in order to detect which ones werepresent in a given situation. The emphasisin the instant report is on gating an active-pixel sensor in the time domain, insteadof filtering light in the wavelength do-main, to prevent the sensor from re-sponding to a laser pulse used to excitefluorescence or phosphorescence whileenabling the sensor to respond to the de-caying fluorescence or phosphorescencesignal that follows the laser pulse. The ac-tive-pixel sensor would be turned on afterthe laser pulse and would be used to ei-ther integrate the fluorescence or phos-phorescence signal over several lifetimesand many excitation pulses or else taketime-resolved measurements of the fluo-rescence or phosphorescence. The reportalso discusses issues of multiplexing andof using time-resolved measurements offluorophores with known different fluo-rescence lifetimes to distinguish amongthem.

This work was done by Gregory Bearmanand Dmitri Kossakovski of Caltech forNASA’s Jet Propulsion Laboratory. Fur-ther information is contained in a TSP (seepage 1).

In accordance with Public Law 96-517, thecontractor has elected to retain title to this inven-tion. Inquiries concerning rights for its commer-cial use should be addressed to

Intellectual Property groupJPLMail Stop 202-2334800 Oak Grove DrivePasadena, CA 91109(818) 354-2240

Refer to NPO-21046, volume and number ofthis NASA Tech Briefs issue, and the pagenumber.

Books & Reports

National Aeronautoics andSpace Adninistration

technology focus computers/electronics - nasa · technology focus computers/electronics software...

Documents