technology focus computers/electronics - nasa immersed in the liquid. ... any gas bubble flowing ......

Technology Focus

Computers/Electronics

Software

Materials

Mechanics

Machinery/Automation

Manufacturing

Bio-Medical

Physical Sciences

Information Sciences

Books and Reports

04-05 April 2005

https://ntrs.nasa.gov/search.jsp?R=20110014845 2018-05-15T15:34:09+00:00Z

NASA Tech Briefs, April 2005 1

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 Innovative Partnerships Offices, 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 CenterLisa L. Lockyer(650) [email protected]

Dryden Flight Research CenterGregory Poteat(661) [email protected]

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

Jet Propulsion LaboratoryKen Wolfenbarger(818) [email protected]

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

Kennedy Space CenterJim Aliberti(321) [email protected]

Langley Research CenterJesse Midgett(757) [email protected]

John H. Glenn Research Center atLewis FieldRobert Lawrence(216) [email protected]

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

Stennis Space CenterJohn Bailey(228) 688-1660 [email protected]

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

Frank SchowengerdtInnovativeTechnology TransferPartnerships (Code TD)(202) [email protected]

John MankinsOffice of Space Flight (Code TD)(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]

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

Gene TrinhHuman Systems Research andTechnology Division (ESMD)(202) [email protected]

John RushSpace Communications Office(SOMD)(202) [email protected]

NASA Field Centers and Program Offices

NNAASSAA PPrrooggrraamm OOffffiicceess

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

5 Technology Focus: Sensors

5 Gas-Tolerant Device Senses Electrical Conductivityof Liquid

5 Nanoactuators Based on Electrostatic Forces onDielectrics

7 Replaceable Microfluidic Cartridges for a PCRBiosensor

7 CdZnTe Image Detectors for Hard-X-RayTelescopes

9 Electronics/Computers

9 High-Aperture-Efficiency Horn Antenna

9 Full-Circle Resolver-to-Linear-Analog Converter

10 Continuous, Full-Circle Arctangent Circuit

11 Advanced Three-Dimensional Display System

13 Software

13 Automatic Focus Adjustment of a Microscope

13 FastScript3D — a Companion to Java 3D

13 Generating Mosaics of Astronomical Images

13 Simulating Descent and Landing of a Spacecraft

14 Simulating Vibrations in a Complex LoadedStructure

14 Rover Sequencing and Visualization Program

14 Software Template for Instruction in Mathematics

14 Support for User Interfaces for DistributedSystems

17 Materials

17 Nanostructured MnO2-Based Cathodes for Li-Ion/Polymer Cells

17 Multi-Layer Laminated Thin Films for InflatableStructures

19 Mechanics

19 Two-Step Laser Ranging for Precise Tracking of a Spacecraft

21 Manufacturing

21 Growing Aligned Carbon Nanotubes forInterconnections in ICs

22 Multilayer Composite Pressure Vessels

23 Bio-Medical

23 Texturing Blood-Glucose-Monitoring Optics UsingOxygen Beams

25 Physical Sciences

25 Fault-Tolerant Heat Exchanger

25 Atomic Clock Based on Opto-Electronic Oscillator

26 Microfocus/Polycapillary-Optic Crystallographic X-Ray System

27 Depth-Penetrating Luminescence Thermographyof Thermal-Barrier Coatings

28 One-Dimensional Photonic Crystal Superprisms

29 Information Sciences

29 Measuring Low-Order Aberrations in aSegmented Telescope

29 Mapping From an Instrumented Glove to a Robot Hand

31 Books & Reports

31 Application of the Hilbert-Huang Transform toFinancial Data

31 Optimizing Parameters for Deep-Space Optical Communication

31 Low-Shear Microencapsulation and ElectrostaticCoating

04-05 April 2005


This document was prepared under the sponsorship of the National Aeronautics and Space Administration. Neither the United States Govern-ment nor any person acting on behalf of the United States Government assumes any liability resulting from the use of the information containedin this document, or warrants that such use will be free from privately owned rights.


Technology Focus: Sensors

The figure depicts a device for measur-ing the electrical conductivity of a flow-ing liquid. Unlike prior such devices, thisone does not trap gas bubbles entrainedin the liquid.

Usually, the electrical conductivity ofa liquid is measured by use of two elec-trodes immersed in the liquid. A typi-cal prior device based on this conceptcontains large cavities that can trapgas. Any gas present between or nearthe electrodes causes a significant off-set in the conductivity reading and, ifthe gas becomes trapped, then the off-set persists.

Extensive tests on two-phase (liq-uid/gas) flow have shown that in thecase of liquid flowing along a section oftubing, gas entrained in the liquid isnot trapped in the section as long asthe inner wall of the section is smoothand continuous, and the section is thenarrowest tubing section along the flowpath. The design of the device is basedon the foregoing observation: Theelectrodes and the insulators separat-ing the electrodes constitute adjacentparts of the walls of a tube. The bore ofthe tube is machined to make the wallsmooth and to provide a straight flowpath from the inlet to the outlet. Thediameter of the electrode/insulatortube assembly is less than the diameter

of the inlet or outlet tubing. An outershell contains the electrodes and insu-lators and constitutes a leak and pres-sure barrier. Any gas bubble flowingthrough this device causes only a mo-mentary conductivity offset that is fil-tered out by software used to processthe conductivity readings.

This work was done by Edward W. O’Con-nor of Hamilton Sundstrand for MarshallSpace Flight Center. For further informa-tion, contact Chris Flynn, Hamilton Sund-strand Company New Technology Representa-tive, at [email protected] to MSC-31931.

A Straight Tube With a Smooth Inner Wall is formed by an assembly of electrodes and insulators in ahollow cylindrical configuration. The electrical conductivity of a liquid flowing along the tube is meas-ured by use of the electrodes.

Insulators

Outer Shell

Electrodes

Gas-Tolerant Device Senses Electrical Conductivity of LiquidBubbles are not trapped in this device.Marshall Space Flight Center, Alabama

Nanoactuators of a proposed typewould exploit the forces exerted by elec-tric fields on dielectric materials. Asused here, “nanoactuators” includes mo-tors, manipulators, and other activemechanisms that have dimensions of theorder of nanometers and/or are de-signed to manipulate objects that havedimensions of the order of nanometers.

The underlying physical principle canbe described most simply in terms of the

example of a square parallel-plate ca-pacitor in which a square dielectricplate is inserted part way into the gapbetween the electrode plates (see Fig-ure 1). Using the conventional approxi-mate equations for the properties of aparallel-plate capacitor, it can readily beshown that the electrostatic field pullsthe dielectric slab toward a central posi-tion in the gap with a force, F, given by

F = V 2(ε1– ε2)a/2d,

where V is the potential applied betweenthe electrode plates, ε1 is the permittivityof the dielectric slab, ε2 is the permittiv-ity of air, a is the length of an electrodeplate, and d is the thickness of the gapbetween the plates.

Typically, the force is small from ourmacroscopic human perspective. Theabove equation shows that the force de-pends on the ratio between the capaci-tor dimensions but does not depend on

Nanoactuators Based on Electrostatic Forces on DielectricsLarge force-to-mass ratios could be achieved at the nanoscale.NASA’s Jet Propulsion Laboratory, Pasadena, California

6 NASA Tech Briefs, April 2005

the size. In other words, the force re-mains the same if the capacitor and thedielectric slab are shrunk to nanometerdimensions. At the same time, themasses of all components are propor-tional to third power of their linear di-mensions. Therefore the force-to-massratio (and, consequently, the accelera-tion that can be imparted to the dielec-tric slab) is much larger at the nano-scale than at the macroscopic scale. Theproposed actuators would exploit thiseffect.

The upper part of Figure 2 depicts asimple linear actuator based on a paral-lel-plate capacitor similar to Figure 1. Inthis case, the upper electrode platewould be split into two parts (A and B)and the dielectric slab would be slightlylonger than plate A or B. The actuatorwould be operated in a cycle. Duringthe first half cycle, plate B would begrounded to the lower plate and plate Awould be charged to a potential, V, withrespect to the lower plate, causing thedielectric slab to be pulled under plateA. During the second half cycle, plate Awould be grounded and plate B wouldbe charged to potential V, causing thedielectric slab to be pulled under plateB. The back-and-forth motion caused byalternation of the voltages on plates Aand B could be used to drive a nano-pump, for example.

A rotary motor, shown in the middlepart of Figure 2, could include a di-electric rotor sandwiched between atop and a bottom plate containingmultiple electrodes arranged symmet-rically in a circle. Voltages would be ap-plied sequentially to electrode pairs 1and 1a, then 2 and 2a, then 3 and 3a inorder to attract the dielectric rotor tosequential positions between the elec-trode pairs.

A micro- or nanomanipulator, shownat the bottom of Figure 2, could in-clude lower and upper plates coveredby rectangular grids of electrodes —in effect, a rectangular array ofnanocapacitors. A dielectric or quasi-dielectric micro- or nanoparticle (abacterium, virus, or molecule for ex-ample) could be moved from an initialposition on the grid to a final positionon the grid by applying a potential se-quentially to the pairs of electrodesalong a path between the initial andfinal positions.

This work was done by Yu Wang of Caltechfor NASA’s Jet Propulsion Laboratory.Further information is contained in a TSP(see page 1).NPO-30747

Figure 2. These Three Devices are examples of nanoactuators that would exploit the principle illus-trated in Figure 1.

A B

RECIPROCATING LINEAR ACTUATOR

ROTARY MOTOR

MICRO- OR NANOMANIPULATOR

A

Electrodes

Rotor

ManipulatedMicro- or Nanoparticle

B

2a3a

1

32

1a

Figure 1. In a Parallel-Plate Capacitor, the electric field pulls a partially inserted dielectric slab furtherinto the gap.

x

a

Force

ElectrodePlates

V

Dielectric Slab

ε2ε1 –

+


Replaceable Microfluidic Cartridges for a PCR BiosensorDesign has been optimized for detection of target DNA sequences. Lyndon B. Johnson Space Center, Houston, Texas

The figure depicts a replaceable mi-crofluidic cartridge that is a componentof a miniature biosensor that detects tar-get deoxyribonucleic acid (DNA) se-quences. The biosensor utilizes (1) poly-merase chain reactions (PCRs) tomultiply the amount of DNA to be de-tected, (2) fluorogenic polynucleotideprobe chemicals for labeling the targetDNA sequences, and (3) a high-sensitiv-ity epifluorescence-detection optoelec-tronic subsystem.

Microfluidics is a relatively new field ofdevice development in which one appliestechniques for fabricating microelectro-mechanical systems (MEMS) to miniaturesystems for containing and/or moving flu-ids. Typically, microfluidic devices are mi-crofabricated, variously, from silicon orpolymers. The development of microflu-idic devices for applications that involvePCR and fluorescence-based detection ofPCR products poses special challenges:• Biocompatibility of materials is a major

requirement. Contact of DNA-contain-ing fluid specimens with silicon inhibitsPCR. It is necessary to either fabricate aPCR microfluidic device from a suit-able polymer or else micromachine thedevice from silicon and coat its interiorsurfaces with a suitable polymer.

• Recently developed polymeric materi-als from which other biocompatiblemicrofluidic devices are made do nothave the high thermal conductivityand low heat capacity needed to facili-tate the rapid thermal cycling that isessential for efficient PCR.

• It is difficult to integrate spectroscopicwindows into microfluidic devices.Thermal-expansion mismatches be-tween silicon substrates and glass win-dows lead to failures of devices.

• Multiple passes of an excitation light

beam are needed to obtain adequatesensitivity for detection; this need fur-ther complicates the problems of de-sign and fabrication.The design and fabrication of the re-

placeable microfluidic cartridges meetsthese challenges. The cartridges aremade from a combination of materials,based on micromachined silicon sub-strates. The thermal-expansion-mis-match problem was solved by use ofthick (3 mm) silicon substrates and an-odically bonded thick (0.7 mm) coversmade of borosilicate float glass. The flu-orescence signal is enhanced by use ofmultipass optics, an essential compo-nent of which is a reflective film ofchemical-vapor-deposited aluminum ona square glass plate affixed to the bottomof the cartridge by use of epoxy. The in-

terior surfaces of the fluidic channelsare coated with dodecyltriethoxysilane.

The design of the cartridge has beenoptimized along with that of the rest ofthe biosensor for detection of targetDNA in a microgravitational or normalgravitational setting. In a test, the biosen-sor was found to be capable of reliablydetecting 600,000 copies of the human β-actin gene after as few as five PCR cycles.

This work was done by Kevin Francis andRon Sullivan of Systems and Processes Engi-neering Corp. for Johnson Space Center.For further information, contact:

SPEC101 West 6th Street, Suite 200Austin, TX 78701-2932Phone: (512) 479-7732Fax: (512) 494-0756E-mail: [email protected]

This Microfluidic Cartridge is a micromachined device that contains fluid specimens during PCR cyclesand serves as an optical cell for detection of fluorogenically labeled DNA sequences.

CdZnTe Image Detectors for Hard-X-Ray TelescopesImage sensors are designed for high spectral resolution and low power consumption.Goddard Space Flight Center, Greenbelt, Maryland

Arrays of CdZnTe photodetectors andassociated electronic circuitry have beenbuilt and tested in a continuing effort todevelop focal-plane image sensor sys-tems for hard-x-ray telescopes. Each

array contains 24 by 44 pixels at a pitchof 498 µm. The detector designs are op-timized to obtain low power demandwith high spectral resolution in the pho-ton-energy range of 5 to 100 keV.

More precisely, each detector array is ahybrid of a CdZnTe photodetector arrayand an application-specific integratedcircuit (ASIC) containing an array of am-plifiers in the same pixel pattern as that


of the detectors. The array is fabricatedon a single crystal of CdZnTe having di-mensions of 23.6 by 12.9 by 2 mm. Thedetector-array cathode is a monolithicplatinum contact. On the anode plane,the contact metal is patterned into theaforementioned pixel array, surroundedby a guard ring that is 1 mm wide onthree sides and is 0.1 mm wide on thefourth side so that two such detector ar-rays can be placed side-by-side to form aroughly square sensor area with minimaldead area between them.

Figure 1 shows two anode patterns.One pattern features larger pixel anodecontacts, with a 30-μm gap betweenthem. The other pattern features smallerpixel anode contacts plus a contact for ashaping electrode in the form of a grid

that separates all the pixels. In operation,the grid is held at a potential intermedi-ate between the cathode and anode po-tentials to steer electric charges towardthe anode in order to reduce the loss ofcharges in the inter-anode gaps.

The CdZnTe photodetector array ismechanically and electrically connectedto the ASIC (see Figure 2), either by useof indium bump bonds or by use of con-ductive epoxy bumps on the CdZnTearray joined to gold bumps on the ASIC.Hence, the output of each pixel detectoris fed to its own amplifier chain.

In the ASIC, each pixel contains a pre-amplifier, a shaping amplifier, a discrim-inator, and sampling and pulsing cir-cuits. All the pixels share a serial readoutline. The ASIC has been designed to op-

erate with low noise and to consume nomore than about 50 mW of power innormal operation.

The ASIC is controlled by a micro-processor — a 24-bit minimum-instruc-tion-set computer (MISC) implementedon a field-programmable gate array. TheMISC runs on a 7.3728-MHz clock cyclethat is established by an oscillator thatruns at a frequency of 14.7456-MHz.Three static random-access memory(SRAM) circuits provide a total of 128kB of 24-bit memory. The output of theASIC readout line is fed to a 12-bit ana-log-to-digital converter (ADC) that con-sumes 80 mW of power. The MISC thenfeeds the output of the ADC to a levelshifter that, in turn, transmits the digitaloutput to an external computer via a se-rial data line.

The sensor system consumes 700 mWof power. By careful design of the ASICand off-chip digital signal processing, ithas been possible to achieve energy res-olution less than 1 keV for hard x-rays atan operating temperature of 0 °C.

This work was done by C. M. Hubert Chen,Walter R. Cook, Fiona A. Harrison, Jiao Y. Y.Lin, Peter H. Mao, and Stephen M. Schindler ofthe California Institute of Technology for God-dard Space Flight Center. Further informa-tion is contained in a TSP (see page 1).

This invention is owned by NASA, and apatent application has been filed. Inquiriesconcerning nonexclusive or exclusive licensefor its commercial development should be ad-dressed to the Patent Counsel, GoddardSpace Flight Center, (301) 286-7351. Referto GSC-14804-1.

Figure 2. An X-Ray Image Sensor System is based on a hybrid of a CdZnTe photodetector array withan ASIC containing an array of amplifier circuits.

CdZnTe Photodetector Array

Bump Bonds

Oscillator

ASIC

MISC

3 × 128-kB SRAM

12-Bit ADC

Serial Line toExternal Computer

Level Shifter

Figure 1. Two Different Anode Patterns have been evaluated: one for a detector array with and one for a detector array without a shaping grid.

1-mm Guard Ring on Three Sides

0.1-mm Guard Ring on Abutting Edge

498-µm Pitch

50-µm Gap From Anode to Grid

14-µm-Wide Grid

300-µm Width for Pixels Along Abutting Edge

Detector Array With Grid

Detector Array Without Grid

1-mm Guard Ring on Three Sides

0.1-mm Guard Ring on Abutting Edge

498-µm Pitch

30-µm Gap From Anode to Grid

300-µm Width for Pixels Along Mating Edge


Electronics/Computers

High-Aperture-Efficiency Horn AntennaMajor design features are a hard (in the electromagnetic sense) boundary and a cosine taper.NASA’s Jet Propulsion Laboratory, Pasadena, California

A horn antenna (see Figure 1) has beendeveloped to satisfy requirements specificto its use as an essential component of ahigh-efficiency Ka-band amplifier: Thecombination of the horn antenna and anassociated microstrip-patch antenna array isrequired to function as a spatial power di-vider that feeds 25 monolithic microwaveintegrated-circuit (MMIC) power ampli-fiers. The foregoing requirement translatesto, among other things, a further require-ment that the horn produce a uniform, ver-tically polarized electromagnetic field in its

aperture in order to feed the microstrippatches identically so that the MMICs canoperate at maximum efficiency.

The horn is fed from a square waveguideof 5.9436-mm-square cross section via atransition piece. The horn features cosine-tapered, dielectric-filled longitudinal cor-rugations in its vertical walls to create ahard boundary condition: This aspect ofthe horn design causes the field in the

horn aperture to be substantially verticallypolarized and to be nearly uniform in am-plitude and phase.

As used here, “cosine-tapered” signifiesthat the depth of the corrugations is a co-sine function of distance along the horn.Preliminary results of finite-element simula-tions of performance have shown that byvirtue of the cosine taper the impedance re-sponse of this horn can be expected to bebetter than has been achieved previously ina similar horn having linearly tapered di-electric-filled longitudinal corrugations.

It is possible to create a hard boundarycondition by use of a single dielectric-filled corrugation in each affected wall,but better results can be obtained with

more corrugations. Simulations were per-formed for a one- and a three-corruga-tion cosine-taper design. For comparison,a simulation was also performed for a lin-ear-taper design (see Figure 2). Thethree-corrugation design was chosen tominimize the cost of fabrication while stillaffording acceptably high performance.Future designs using more corrugationsper wavelength are expected to providebetter field responses and, hence, greateraperture efficiencies.

This work was done by Wesley Pickens,Daniel Hoppe, Larry Epp, and Abdur Kahnof Caltech for NASA’s Jet Propulsion Labo-ratory. Further information is contained in aTSP (see page 1). NPO-40023

Figure 1. This Horn Antenna features cosine-ta-pered corrugation that imparts desired imped-ance characteristics.

Figure 2. The Computed Input Reflection Coefficient in the desired electromagnetic mode was found tobe more nearly constant with frequency for the cosine-taper designs than for the linear-taper design.The maximum depth of the taper [73 mils (1.85 mm)] is the same in all three designs.

0.0

–10.0

–20.0

–30.031.80 31.90 32.00 32.10 32.20 32.30

Am

plit

ud

e, d

B

Three Corrugations,Cosine Taper

One Corrugation,Cosine Taper

One Corrugation,Linear Taper

Frequency, GHz

Full-Circle Resolver-to-Linear-Analog ConverterThis circuit costs less and is less susceptible to error.Marshall Space Flight Center, Alabama

A circuit generates sinusoidal excita-tion signals for a shaft-angle resolverand, like the arctangent circuit de-scribed in the preceding article, gener-ates an analog voltage proportional to

the shaft angle. The disadvantages ofthe circuit described in the precedingarticle arise from the fact that it must bemade from precise analog subcircuits,including a functional block capable of

implementing some trigonometricidentities; this circuitry tends to be ex-pensive, sensitive to noise, and suscepti-ble to errors caused by temperature-in-duced drifts and imprecise matching of


ExcitationCircuit

Kcos(ωt)Kcos(ωt)

Ksin(ωt)

Ground

InvertingSquarer

–LL[cos(ωt)]

Unity-GainBuffer

Kcos(ωt–Θ)

Kcos(ωt–Θ)

Ksin(ωt–Θ)

Ground

InvertingSquarer

Amplifierand Level

Shifter

–LL[cos(ωt–Θ)]

Block ofDigitalLogic

Circuits

PWM(Θ)

0 to 5 VDCLow-Pass

Filter

PWM(Θ)

–10 to +10 VDC

OutputVoltage

Proportionalto Θ

To ExcitationTerminals ofShaft-Angle

Resolver

From OutputTerminals ofShaft-Angle

Resolver

The Resolver-to-Linear-Analog Converter is depicted here in simplified form. This circuit provides excitation for a shaft-angle resolver and generates a DCoutput voltage proportional to the shaft angle, Θ.

Continuous, Full-Circle Arctangent CircuitThe discontinuity of the tangent function at 90° causes no trouble.Marshall Space Flight Center, Alabama

A circuit generates an analog voltageproportional to an angle, in response totwo sinusoidal input voltages havingmagnitudes proportional to the sineand cosine of the angle, respectively.That is to say, given input voltages pro-portional to sin(ωt)sin(Θ) andsin(ωt)cos(Θ) [where Θ denotes theangle, ω denotes 2π × a carrier fre-quency, and t denotes time], the circuitgenerates a steady voltage proportionalto Θ. The output voltage varies continu-ously from its minimum to its maxi-mum value as Θ varies from –180° to180°. While the circuit could acceptinput modulated sine and cosine sig-nals from any source, it must be notedthat such signals are typical of the out-puts of shaft-angle resolvers in electro-magnetic actuators used to measureand control shaft angles for diverse pur-poses like aiming scientific instrumentsand adjusting valve openings.

In effect, the circuit is an analog com-puter that calculates the arctangent of theratio between the sine and cosine signals.The full-circle angular range of this arctan-gent circuit stands in contrast to the rangeof prior analog arctangent circuits, which isfrom slightly greater than –90° to slightlyless than +90°. Moreover, for applicationsin which continuous variation of output ispreferred to discrete increments of output,this circuit offers a clear advantage over re-solver-to-digital integrated circuits.

The figure depicts the main func-tional blocks of the arctangent circuit.In addition to the aforementioned inputsignals proportional to sin(ωt)sin(Θ)and sin(ωt)cos(Θ), the circuit receivesthe carrier signal proportional to sin(ωt)as an auxiliary input. The carrier signalis fed to a squarer (block 7) to obtain anoutput square-wave or logic-level signal,LL[sin(ωt)]. The demodulator (block1) uses LL[sin(ωt)] to demodulate input

signal sin(ωt)cos(Θ), generating an out-put proportional to cos(Θ).

The carrier signal sin(ωt) is also fed toan integrator and inverter (block 8) toobtain a signal proportional to cos(ωt).The cos(ωt) signal is fed to a squarer(block 9) to obtain a logic-level signalLL[cos(ωt)]. The cos(Θ) and cos(ωt)signals are fed to a multiplier (block 2)to obtain a signal proportional tocos(Θ)cos(ωt). This signal and the inputsin(ωt)sin(Θ) signal are fed to an in-verter and adder (block 3) to obtain asignal proportional to –[cos(Θ)cos(ωt)+ sin(Θ)sin(ωt)], which, by trigonomet-ric identity, equals –cos(ωt – Θ). This sig-nal is processed by an inverter andsquarer (block 4) to obtain a logic-levelsignal LL[cos(ωt – Θ)].

The signal LL[cos(ωt)] from block 9and the signal LL[cos(ωt – Θ)] fromblock 4 have the same frequency but dif-fer in phase by Θ. These signals are fed

gains and phases. These disadvantagesare overcome by the design of the pres-ent circuit.

The present circuit (see figure) in-cludes an excitation circuit, which gener-ates signals K sin(ωt) and Kcos(ωt)[where K is an amplitude, ω denotes 2π ×a carrier frequency (the design value ofwhich is 10 kHz), and t denotes time].These signals are applied to the excita-tion terminals of a shaft-angle resolver,causing the resolver to put out signals Csin(ωt – Θ) and C cos(ωt – Θ). The cosineexcitation signal and the cosine resolveroutput signal are processed through in-verting comparator circuits, which areconfigured to function as inverting squar-

ers, to obtain logic-level or square-wavesignals –LL[cos(ωt)] and –LL[cos(ωt –Θ)], respectively. These signals are fed asinputs to a block containing digital logiccircuits that effectively measure the phasedifference (which equals Θ between thetwo logic-level signals). The output of thisblock is a pulse-width-modulated signal,PWM(Θ), the time-averaged value ofwhich ranges from 0 to 5 VDC as Θranges from –180 to +180°.

PWM(Θ) is fed to a block of amplify-ing and level-shifting circuitry, whichconverts the input PWM waveform to anoutput waveform that switches betweenprecise reference voltage levels of +10and –10 V. This waveform is processed

by a two-pole, low-pass filter, which re-moves the carrier-frequency compo-nent. The final output signal is a DC po-tential, proportional to Θ that rangescontinuously from –10 V at Θ = –180° to+10 V at Θ = +180°.

This work was done by Dean C. Alhorn,Dennis A. Smith, and David E. Howard ofMarshall Space Flight Center.

This invention has been patented by NASA(U.S. Patent No. 6,104,328). Inquiries con-cerning nonexclusive or exclusive license forits commercial development should be ad-dressed to Sammy Nabors, MSFC Commer-cialization Assistance Lead, at (256) 544-5226 or [email protected]. Refer toMFS-31237.


as inputs to block 5, which contains logiccircuitry that determines the magnitudeand trigonometric quadrant of thephase difference, and generates a logic-level pulse-width-modulated signal,PWM(Θ), in which the pulse widthvaries continuously with Θ. The quad-rant-detection function eliminates thedifficulty, encountered in prior analogarctangent circuits, caused by the dis-continuity of the tan(Θ) at Θ = ±90°.

PWM(Θ) is fed to block 6, which re-sponds by generating a PWM waveformthat switches between precise referencevoltage levels of +10 and –10 V. Thiswaveform is processed by a two-pole,low-pass filter (block 10), which filtersout the carrier-frequency component.The output of block 10 is a DC potential,proportional to Θ, that ranges continu-ously from –10 V at Θ = –180° to +10 V atΘ = +180°.

This work was done by Dean C. Alhorn,David E. Howard, and Dennis A. Smith ofMarshall Space Flight Center.

This invention has been patented by NASA(U.S. Patent No. 6,138,131). Inquiries con-cerning nonexclusive or exclusive license forits commercial development should be ad-dressed to Sammy Nabors, MSFC Commer-cialization Assistance Lead, at (256) 544-5226 or [email protected]. Refer toMFS-31219.

Demodulator(Block 1)

Multiplier(Block 2)

Squarer(Block 9)

Two-Pole Filter(Block 10)

Buffer(Block 11)

Squarer(Block 7)

Inverter & Integrator(Block 8)

Inverter & Adder(Block 3)

Inverter & Squarer(Block 4)

±10 Volt ReferenceWith Switching

(Block 6)

Logic Circuit to Developa Pulse-Width-ModulatedSignal & Determine the

Quadrant (Block 5)

sin(ωt)

sin(ωt)cos(Θ) cos(Θ) cos(Θ)cos(ωt) –cos(ωt – Θ)

sin(ωt)sin(Θ)

cos(ωt)

LL[sin(ωt)]

LL[cos(ωt – Θ)]

LL[cos(ωt)]

Logic Level PWM (Θ)

±10 Volt LinearRepresentation of Θ

Four QuadrantContinuous 360°

Linear Θ

Output(± 10 Volt)

±10 Volt PWM (Θ)

This Circuit Generates a DC Voltage Proportional to Θ in response to input voltages proportional to sin(ωt)sin(Θ) and sin(ωt)cos(Θ) and an auxiliary inputvoltage proportional to sin(ωt).

A desktop-scale, computer-controlleddisplay system, initially developed forNASA and now known as the Vol-umeViewer®, generates three-dimen-sional (3D) images of 3D objects in adisplay volume. This system differs fun-damentally from stereoscopic and holo-graphic display systems: The imagesgenerated by this system are truly 3D inthat they can be viewed from almost anyangle, without the aid of special eye-glasses. It is possible to walk around thesystem while gazing at its display volumeto see a displayed object from a chang-ing perspective, and multiple observersstanding at different positions aroundthe display can view the object simulta-neously from their individual perspec-

tives, as though the displayed objectwere a real 3D object.

At the time of writing this article, onlypartial information on the design andprinciple of operation of the system wasavailable. It is known that the system in-cludes a high-speed, silicon-backplane,ferroelectric-liquid-crystal spatial lightmodulator (SLM), multiple high-powerlasers for projecting images in multiplecolors, a rotating helix that serves as amoving screen for displaying voxels [vol-ume cells or volume elements, in anal-ogy to pixels (picture cells or picture el-ements) in two-dimensional (2D)images], and a host computer. The rotat-ing helix and its motor drive are the onlymoving parts. Under control by the host

computer, a stream of 2D image patternsis generated on the SLM and projectedthrough optics onto the surface of therotating helix.

The system utilizes a parallelpixel/voxel-addressing scheme: All thepixels of the 2D pattern on the SLM areaddressed simultaneously by laserbeams. This parallel addressing schemeovercomes the difficulty of achievingboth high resolution and a high framerate in a raster scanning or serial ad-dressing scheme.

It has been reported that the structureof the system is simple and easy to build,that the optical design and alignment arenot difficult, and that the system can bebuilt by use of commercial off-the-shelf

Advanced Three-Dimensional Display SystemThe display can be viewed from almost any direction, without special eyeglasses.Stennis Space Center, Mississippi


products. A prototype of the system dis-plays an image of 1,024 by 768 by 170(=133,693,440) voxels. In future designs,the resolution could be increased. Themaximum number of voxels that can begenerated depends upon the spatial res-olution of SLM and the speed of rotationof the helix. For example, one could usean available SLM that has 1,024 by 1,024pixels. Incidentally, this SLM is capable

of operation at a switching speed of300,000 frames per second.

Implementation of full-color displaysin future versions of the system would bestraightforward: One could use threeSLMs for red, green, and blue, respec-tively, and the colors of the voxels couldbe automatically controlled. An opticallysimpler alternative would be to use a sin-gle red/green/ blue light projector and

synchronize the projection of each colorwith the generation of patterns for thatcolor on a single SLM.

This work was done by Jason Geng of GenexTechnologies, Inc., for Stennis Space Center.

Inquiries concerning rights for the commercialuse of this invention should be addressed to the In-tellectual Property Manager, Stennis Space Center,(228) 688-1929. Refer to SSC-00205.


Software

Automatic Focus Adjust-ment of a Microscope

AUTOFOCUS is a computer programfor use in a control system that automat-ically adjusts the position of an instru-ment arm that carries a microscopeequipped with an electronic camera. Inthe original intended application ofAUTOFOCUS, the imaging microscopewould be carried by an exploratory ro-botic vehicle on a remote planet, butAUTOFOCUS could also be adapted tosimilar applications on Earth. Initiallycontrol software other than AUTOFO-CUS brings the microscope to a positionabove a target to be imaged. Then theinstrument arm is moved to lower themicroscope toward the target: nomi-nally, the target is approached from astarting distance of 3 cm in 10 steps of 3mm each. After each step, the image inthe camera is subjected to a wavelettransform, which is used to evaluate thetexture in the image at multiple scales todetermine whether and by how muchthe microscope is approaching focus. Afocus measure is derived from the trans-form and used to guide the arm to bringthe microscope to the focal height.When the analysis reveals that the micro-scope is in focus, image data arerecorded and transmitted.

This program was written by TerranceHuntsberger of Caltech for NASA’s JetPropulsion Laboratory. Further informa-tion is contained in a TSP (see page 1).

This software is available for commerciallicensing. Please contact Karina Edmonds ofthe California Institute of Technology at(818) 393-2827. Refer to NPO-30531.

FastScript3D — a Companion to Java 3D

FastScript3D is a computer program,written in the Java 3D™ programminglanguage, that establishes an alternativelanguage that helps users who lack ex-pertise in Java 3D to use Java 3D for con-structing three-dimensional (3D)-ap-pearing graphics. The FastScript3Dlanguage provides a set of simple, intu-itive, one-line text-string commands forcreating, controlling, and animating 3Dmodels. The first word in a string is thename of a command; the rest of thestring contains the data arguments forthe command. The commands can also

be used as an aid to learning Java 3D. De-velopers can extend the language byadding custom text-string commands.The commands can define new 3D ob-jects or load representations of 3D ob-jects from files in formats compatiblewith such other software systems as X3D.The text strings can be easily integratedinto other languages. FastScript3D facili-tates communication between scriptinglanguages [which enable programmingof hyper-text markup language (HTML)documents to interact with users] andJava 3D. The FastScript3D language canbe extended and customized on both thescripting side and the Java 3D side.

This program was written by Patti Koenigof Caltech for NASA’s Jet Propulsion Lab-oratory. Further information is containedin a TSP (see page 1).


Generating Mosaics of Astronomical Images

“Montage” is the name of a service ofthe National Virtual Observatory (NVO),and of software being developed to imple-ment the service via the World Wide Web.Montage generates science-grade custommosaics of astronomical images on de-mand from input files that comply withthe Flexible Image Transport System(FITS) standard and contain image dataregistered on projections that complywith the World Coordinate System (WCS)standards. “Science-grade” in this contextsignifies that terrestrial and instrumentalfeatures are removed from images in away that can be described quantitatively.“Custom” refers to user-specified parame-ters of projection, coordinates, size, rota-tion, and spatial sampling. The greatestvalue of Montage is expected to lie in itsability to analyze images at multiple wave-lengths, delivering them on a commonprojection, coordinate system, and spatialsampling, and thereby enabling furtheranalysis as though they were part of a sin-gle, multi-wavelength image. Montage willbe deployed as a computation-intensiveservice through existing astronomy por-tals and other Web sites. It will be inte-grated into the emerging NVO architec-ture and will be executed on theTeraGrid. The Montage software will also

be portable and publicly available.This program was written by Attila Bergou,

Bruce Berriman, John Good, Joseph Jacob,Daniel Katz, Anastasia Laity, Thomas Prince,and Roy Williams of Caltech for NASA’s JetPropulsion Laboratory. Further informa-tion is contained in a TSP (see page 1).


Simulating Descent andLanding of a Spacecraft

The Dynamics Simulator for Entry, De-scent, and Surface landing (DSENDS) soft-ware performs high-fidelity simulation ofthe Entry, Descent, and Landing (EDL) ofa spacecraft into the atmosphere and ontothe surface of a planet or a smaller body.DSENDS is an extension of the DShell andDARTS programs (described in priorNASA Tech Briefs articles), which afford ca-pabilities for mathematical modeling of thedynamics of a spacecraft as a whole and ofits instruments, actuators, and other subsys-tems. DSENDS enables the modeling (in-cluding real-time simulation) of flight-trainelements and all spacecraft responses dur-ing various phases of EDL. DSENDS pro-vides high-fidelity models of the aerody-namics of entry bodies and parachutes plussupporting models of atmospheres. Terrainand real-time responses of terrain-imagingradar and lidar instruments can also bemodeled. The program includes modulesfor simulation of guidance, navigation, hy-personic steering, and powered descent.Automated state-machine-driven modelswitching is used to represent spacecraftseparations and reconfigurations. Modelsfor computing landing contact and impactforces are expected to be added. DSENDScan be used as a stand-alone program or in-corporated into a larger program that sim-ulates operations in real time.

This program was written by J. Balaram, Ab-hinandan Jain, Bryan Martin, ChristopherLim, David Henriquez, Elihu McMahon, GarettSohl, Pranab Banerjee, Robert Steele, and Timo-thy Bentley of Caltech, and Scott Striepe and BrettStarr of Langley Research Center for NASA’s JetPropulsion Laboratory. Further informationis contained in a TSP (see page 1).



Simulating Vibrations in aComplex Loaded Structure

The Dynamic Response Computation(DIRECT) computer program simulatesvibrations induced in a complex structureby applied dynamic loads. Developed toenable rapid analysis of launch- and land-ing-induced vibrations and stresses in aspace shuttle, DIRECT also can be usedto analyze dynamic responses of otherstructures — for example, the response ofa building to an earthquake, or the re-sponse of an oil-drilling platform and at-tached tanks to large ocean waves. For aspace-shuttle simulation, the requiredinput to DIRECT includes mathematicalmodels of the space shuttle and its pay-loads, and a set of forcing functions thatsimulates launch and landing loads. DI-RECT can accommodate multiple levelsof payload attachment and substructureas well as nonlinear dynamic responses ofstructural interfaces. DIRECT combinesthe shuttle and payload models into a sin-gle structural model, to which the forcingfunctions are then applied. The resultingequations of motion are reduced to anoptimum set and decoupled into aunique format for simulating dynamics.During the simulation, maximum vibra-tions, loads, and stresses are monitoredand recorded for subsequent analysis toidentify structural deficiencies in the shut-tle and/or payloads.

This program was written by Tim T. Cao ofJohnson Space Center. For further informa-tion, contact the Johnson Commercial Technol-ogy Office at (281) 483-3809.MSC-23333

Rover Sequencing and Visualization Program

The Rover Sequencing and Visualiza-tion Program (RSVP) is the software toolfor use in the Mars Exploration Rover(MER) mission for planning rover oper-ations and generating command se-quences for accomplishing those opera-tions. RSVP combines three-dimensional(3D) visualization for immersive explo-ration of the operations area, stereo-scopic image display for high-resolutionexamination of the downlinked imagery,and a sophisticated command-sequenceediting tool for analysis and completionof the sequences. RSVP is linked with ac-tual flight-code modules for operationsrehearsal to provide feedback on the ex-pected behavior of the rover prior tocommitting to a particular sequence.Playback tools allow for review of both re-

hearsed rover behavior and downlinkedresults of actual rover operations. Thesecan be displayed simultaneously for com-parison of rehearsed and actual activitiesfor verification.

The primary inputs to RSVP are down-link data products from the OperationsStorage Server (OSS) and activity plansgenerated by the science team. The activ-ity plans are high-level goals for the nextday’s activities. The downlink data prod-ucts include imagery, terrain models, andtelemetered engineering data on roveractivities and state. The Rover SequenceEditor (RoSE) component of RSVP per-forms activity expansion to command se-quences, command creation and editingwith setting of command parameters, andviewing and management of rover re-sources. The HyperDrive component ofRSVP performs 2D and 3D visualizationof the rover’s environment, graphical andanimated review of rover-predicted andtelemetered state, and creation and edit-ing of command sequences related tomobility and Instrument Deployment De-vice (IDD) operations. Additionally,RoSE and HyperDrive together evaluatecommand sequences for potential viola-tions of flight and safety rules. The prod-ucts of RSVP include command se-quences for uplink that are stored in theDistributed Object Manager (DOM) andpredicted rover state histories stored inthe OSS for comparison and validation ofdownlinked telemetry.

The majority of components compris-ing RSVP utilize the MER command andactivity dictionaries to automatically cus-tomize the system for MER activities. Thus,RSVP, being highly data driven, may be tai-lored to other missions with minimal ef-fort. In addition, RSVP uses a distributed,message-passing architecture to allow mul-titasking, and collaborative visualizationand sequence development by scatteredteam members.

This tool was developed by Brian Cooper,Frank Hartman, Scott Maxwell, Jeng Yen,John Wright, and Carlos Balacuit of Cal-tech for NASA’s Jet Propulsion Labora-tory. Further information is contained in aTSP (see page 1).


Software Template for Instruction in Mathematics

Intelligent Math Tutor (IMT) is a soft-ware system that serves as a template for

creating software for teaching mathe-matics. IMT can be easily connected toartificial-intelligence software and otheranalysis software through input and out-put of files. IMT provides an easy-to-useinterface for generating courses that in-clude tests that contain both multiple-choice and fill-in-the-blank questions,and enables tracking of test scores. IMTmakes it easy to generate software forWeb-based courses or to manufacturecompact disks containing executablecourse software. IMT also can functionas a Web-based application program,with features that run quickly on theWeb, while retaining the intelligence ofa high-level language application pro-gram with many graphics. IMT can beused to write application programs intext, graphics, and/or sound, so thatthe programs can be tailored to theneeds of most handicapped persons.The course software generated by IMTfollows a “back to basics” approach ofteaching mathematics by inducing thestudent to apply creative mathematicaltechniques in the process of learning.Students are thereby made to discovermathematical fundamentals andthereby come to understand mathemat-ics more deeply than they couldthrough simple memorization.

This program was written by Robert O.Shelton of Johnson Space Center, andTravis A. Moebes and Anna Beall of ScienceApplications International Corp. For furtherinformation, contact the Johnson CommercialTechnology Office at (281) 483-3809.MSC-23614

Support for User Interfacesfor Distributed Systems

An extensible Java™ software frame-work supports the construction and op-eration of graphical user interfaces(GUIs) for distributed computing sys-tems typified by ground control systemsthat send commands to, and receivetelemetric data from, spacecraft.Heretofore, such GUIs have been cus-tom built for each new system at consid-erable expense. In contrast, the presentframework affords generic capabilitiesthat can be shared by different distrib-uted systems. Dynamic class loading, re-flection, and other run-time capabilitiesof the Java language and JavaBeanscomponent architecture enable the cre-ation of a GUI for each new distributedcomputing system with a minimum ofcustom effort. By use of this framework,GUI components in control panels and


menus can send commands to a partic-ular distributed system with a minimumof system-specific code. The frameworkreceives, decodes, processes, and dis-plays telemetry data; custom telemetrydata handling can be added for a par-ticular system. The framework supportssaving and later restoration of users’

configurations of control panels andtelemetry displays with a minimum ofeffort in writing system-specific code.GUIs constructed within this frame-work can be deployed in any operatingsystem with a Java run-time environ-ment, without recompilation or codechanges.

This program was written by Glenn Eychanerand Albert Niessner of Caltech for NASA’s JetPropulsion Laboratory. Further informa-tion is contained in a TSP (see page 1).



Materials

Multi-Layer Laminated Thin Films for Inflatable StructuresLaminates offer advantages over equal-thickness monolayer sheets.NASA’s Jet Propulsion Laboratory, Pasadena, California

Special-purpose balloons and other in-flatable structures would be constructedas flexible laminates of multiple thinpolymeric films interspersed with layersof adhesive, according to a proposal. Inthe original intended application, thelaminate would serve as the envelope ofthe Titan Aerobot — a proposed roboticairship for exploring Titan (one of themoons of Saturn). Potential terrestrialapplications for such flexible laminatescould include blimps and sails.

In the original application, the multi-lay-ered laminate would contain six layers of0.14-mil (0.0036-mm)-thick Mylar®‚ (orequivalent) polyethylene terephthalate filmwith a layer of adhesive between each layerof Mylar®. The overall thickness and arealdensity of this laminate would be nearly thesame as those of 1-mil (0.0254-mm)-thickmonolayer polyethylene terephthalatesheet. However, the laminate would offerseveral advantages over the monolayersheet, especially with respect to interrelated

considerations of flexing properties, forma-tion of pinholes, and difficulty or ease ofhandling, as discussed next.

Most of the damage during flexing ofthe laminate would be localized in theoutermost layers, where the radii ofbending in a given bend would be thelargest and, hence, the bending stresswould be the greatest. The adverse ef-fects of formation of pinholes would benearly completely mitigated in the lami-nate because a pinhole in a given layer

Nanostructured MnO2-Based Cathodes for Li-Ion/Polymer CellsExperiments show promise for increasing energy densities.Lyndon B. Johnson Space Center, Houston, Texas

Nanostructured MnO2-based cath-odes for Li-ion/polymer electrochemi-cal cells have been investigated in a continuing effort to develop safe, high-energy-density, reliable, low-toxicity,rechargeable batteries for a variety of ap-plications in NASA programs and inmass-produced commercial electronicequipment. Whereas the energy densi-ties of state-of-the-art lithium-ion/poly-mer batteries range from 150 to 175

W⋅h/kg, the goal of this effort is to in-crease the typical energy density toabout 250 W⋅h/kg. It is also expectedthat an incidental benefit of this effortwill be increases in power densities be-cause the distances over which Li ionsmust diffuse through nanostructuredcathode materials are smaller than thosethrough solid bulk cathode materials.

The developmental MnO2-based cath-ode nanostructures are, more specifically,

layered structures thathave compositions ofLixMn1–yMyO2 (wherex≥0, 0≤y≤1, and M de-notes a dopant metalother than Mn or Li).The average size ofcrystallites in cathodesmade from nanolay-ered LixMn1–yMyO2was observed to beabout 50 nm. The en-ergy densities ofthese cathodes werefound to be approxi-mately 440 W⋅h/kg.The charge/dischargecurves of these cath-odes were observed tolie in the potentialrange of 4.2 to 2 V andto be continuous. The

nanostructures were found to be stableand the charge/discharge capacities ofthe cathodes were found not to fadeafter multiple charge/discharge cycles(see figure).

The experiments performed thus farhave involved small laboratory cells.Further research will be needed todemonstrate practical cells. The tailor-ing of nanostructures and compositionsis likely to be an important topic of re-search because the electrochemicalproperties of LixMn1–yMyO2 from whichthe cathodes are made depend on thesizes of the crystallites, and the type andthe amounts of dopants.

This work was done by Ganesh Skandanand Amit Singhal of Nanopowder EnterprisesInc. for Johnson Space Center.

In accordance with Public Law 96-517,the contractor has elected to retain title to thisinvention. Inquiries concerning rights for itscommercial use should be addressed to:

Dr. Ganesh SkandanNanopowder Enterprises Inc.120 Centennial AvenuePiscataway, NJ 08854-3908Phone: (732) 885-1088E-mail:ganeshskandan@

nanopowderenterprises.comRefer to MSC-23368, volume and number

of this NASA Tech Briefs issue, and thepage number.

00

50

100

150

200

5 10

Cycle Number

Cap

acit

y, m

A·h

/g

15 20

The Charge/Discharge Capacities of LixMn1–yMyO2 cathode materialswere observed not to decrease below initial values after 20 charge/dis-charge cycles.


would not propagate to adjacent layers.Hence, the laminate would tend to re-main effective as a barrier to retain gas.Similar arguments can be made regard-ing cracks: While a crack could form as aresult of stress or a defect in the film ma-terial, a crack would not propagate intoadjacent layers, and the adjacent

layer(s) would even arrest propagationof the crack.

In the case of the monolayer sheet,surface damage (scratches, dents, per-manent folds, pinholes, and the like)caused by handling would constitute orgive rise to defects that could propa-gate through the thickness as cracks orpinholes that would render the sheetless effective or ineffective as a barrier.In contrast, because damage incurredduring handling of the laminate wouldordinarily be limited to the outermostlayers, the barrier properties of thelaminate would be less likely to be ad-versely affected. Therefore, handlingof the laminate would be easier be-cause there would be less of a need toexercise care to ensure against surfacedamage.

For the Titan Aerobot, the laminate isrequired to retain its physical properties(especially flexibility and effectiveness asa barrier) to a sufficient degree at tem-peratures as low as that of liquid nitro-gen. To evaluate this laminate and othercandidate materials, a flex testing appa-ratus (see figure) has been used to re-peatedly flex samples of the materialswith a 45° twist and a 2-in. (≈5-cm) com-pression while the samples were im-mersed in liquid nitrogen. After having

been flexed a set number of cycles, sam-ples were examined by use of an appara-tus that can easily detect gas leaks fromthrough pinholes as narrow as 10 μm indiameter. In this test, a six-layer polyeth-ylene terephthalate laminate as de-scribed above survived more than 3,400flex cycles in liquid nitrogen without de-veloping through pinholes — perform-ing significantly better than did a mono-layer polyethylene terephthalate sheetof equivalent overall thickness.

To evaluate these materials for utility asterrestrial balloon materials, the flexingand pinhole tests were performed atroom temperature. As in the liquid-nitro-gen tests, the laminate performed betterthan did the monolayer sheet.

In a contemplated improvement on thebasic laminate design, a layer (or layers) ofreinforcing fabric would be laminatedwith the layers of polymeric film and layersof adhesive. At the time of reporting theinformation for this article, evaluation ofcandidate materials for use in such fabric-augmented laminates was in progress.

This work was done by Andre Yavrouian,Gary Plett, and Jerami Mannella of Cal-tech for NASA’s Jet Propulsion Labora-tory. Further information is contained ina TSP (see page 1).NPO-40636

A Flex Testing Apparatus repeatedly twists andcompresses a sample of material.


Mechanics

Two-Step Laser Ranging for Precise Tracking of a SpacecraftNASA’s Jet Propulsion Laboratory, Pasadena, California

A document proposes a two-step laserranging technique for precise trackingof a coasting interplanetary spacecraftto determine the degree to which leak-age of fuel, solar wind, and/or solar-ra-diation pressure causes it to deviatefrom a purely gravitational trajectory.Such a determination could contributeto the precision of a test of a theory ofgravitation. In the technique, a proofmass would be released from the space-craft. By use of laser ranging equipment

on the spacecraft and retroreflectors at-tached to the proof mass, the relativeposition of the spacecraft and proofmass would be determined. Meanwhile,the position of the spacecraft relative tothe Earth would be determined byranging by use of a laser transponder.The vector sum of the two sets of rang-ing measurements would be the posi-tion of the proof mass relative to theEarth. Unlike the acceleration of thespacecraft, the acceleration of the

proof mass should not include a resid-ual component attributable to leakageof fuel. In addition, the effects of solarradiation and solar wind on the proofmass could be minimized by releasingthe proof mass into the shadow of thespacecraft.

This work was done by Talso Chui and Kon-stantin Penanen of Caltech for NASA’s JetPropulsion Laboratory. Further informa-tion is contained in a TSP (see page 1).NPO-40733


Manufacturing

A process for growing multiwalledcarbon nanotubes anchored at specifiedlocations and aligned along specified di-rections has been invented. Typically,one would grow a number of the nan-otubes oriented perpendicularly to a sil-

icon integrated-circuit (IC) sub-strate, starting from (and an-chored on) patterned catalyticspots on the substrate. Such ar-rays of perpendicular carbonnanotubes could be used as elec-trical interconnections betweenlevels of multilevel ICs.

The process (see Figure 1) be-gins with the formation of alayer, a few hundred nanometersthick, of a compatible electri-cally insulating material (e.g.,SiOx or SiyNz) on the silicon sub-strate. A patterned film of a suit-able electrical conductor (Al,Mo, Cr, Ti, Ta, Pt, Ir, or dopedSi), having a thickness between 1nm and 2 µm, is deposited onthe insulating layer to form theIC conductor pattern. Next, acatalytic material (usually, Ni,Fe, or Co) is deposited to a thick-ness between 1 and 30 nm onthe spots from which it is desiredto grow carbon nanotubes.

The carbon nanotubes aregrown by plasma-enhanced chem-ical vapor deposition (PECVD).Unlike the matted and tangledcarbon nanotubes grown by ther-mal CVD, the carbon nanotubesgrown by PECVD are perpendicu-lar and freestanding because an electricfield perpendicular to the substrate isused in PECVD. Next, the free space be-tween the carbon nanotubes is filled withSiO2 by means of CVD from tetraethy-lorthosilicate (TEOS), thereby forming anarray of carbon nanotubes embedded inSiO2. Chemical mechanical polishing(CMP) is then performed to remove ex-cess SiO2 and form a flat-top surface inwhich the outer ends of the carbon nan-otubes are exposed. Optionally, depend-ing on the application, metal lines to con-nect selected ends of carbon nanotubesmay be deposited on the top surface.

The top part of Figure 2 is a scanningelectron micrograph (SEM) of carbonnanotubes grown, as described above,on catalytic spots of about 100 nm diam-eter patterned by electron-beam lithog-

raphy. These and other nanotubes werefound to have lengths ranging from 2 to10 µm and diameters ranging from 30to 200 nm, the exact values of length de-pending on growth times and condi-tions and the exact values of diameterdepending on the diameters and thick-nesses of the catalyst spots. The bottonpart of Figure 2 is an SEM of an embed-ded array of carbon nanotubes afterCMP.

This work was done by Jun Li, Qi Ye, AlanCassell, Hou Tee Ng, Ramsey Stevens, Jie Han,and M. Meyyappan of Ames Research Cen-ter. Further information is contained in a TSP(see page 1).

Inquiries concerning rights for the commer-cial use of this invention should be addressedto the Patent Counsel, Ames Research Center,(650) 604-5104. Refer to ARC-15042-1.

Growing Aligned Carbon Nanotubes for Interconnections in ICsCarbon nanotubes are embedded in silica with their tips exposed as contacts. Ames Research Center, Moffett Field, California

Deposition of Metal

Deposition and Patterning of Catalyst

PECVDCarbon Nanotubes

Substrate With Top Insulating Layer

TEOSCVD

CMP

Deposition of Metal on Tips of Nanotubes

Figure 1. A Bottom-Up Approach is followed ingrowing carbon nanotubes, embedding them ina silica matrix, exposing their tips, and using theexposed tips as electrical contacts.

Figure 2. These Electron Micrographs, both taken at anangle of 45°, show specimens of arrays of carbon nan-otubes at two different stages of processing. (The scale baron the top is 5 μm, the one on the bottom is 0.2 μm.)


A method has been devised to en-able the fabrication of lightweightpressure vessels from multilayer com-posite materials. This method is re-lated to, but not the same as, themethod described in “Making a Metal-Lined Composite-Overwrapped Pres-sure Vessel” (MFS-31814), NASA TechBriefs, Vol. 29, No. 3 (March 2005),page 59. The method is flexible in thatit poses no major impediment tochanges in tank design and is applica-ble to a wide range of tank sizes.

The figure depicts a finished tank fab-ricated by this method, showing layersadded at various stages of the fabrica-tion process. In the first step of theprocess, a mandrel that defines the sizeand shape of the interior of the tank ismachined from a polyurethane foam orother suitable lightweight tooling mate-rial. The mandrel is outfitted withmetallic end fittings on a shaft. Eachend fitting includes an outer flange thathas a small step to accommodate a thinlayer of graphite/epoxy or other suit-able composite material. The outer sur-face of the mandrel (but not the fit-tings) is covered with a suitable releasematerial. The composite material is fila-ment-wound so as to cover the entiresurface of the mandrel from the step onone end fitting to the step on the otherend fitting. The composite material isthen cured in place.

The entire workpiece is cut in half ina plane perpendicular to the axis ofsymmetry at its mid-length point, yield-ing two composite-material half shells,each containing half of the foam man-drel. The halves of the mandrel are re-moved from within the compositeshells, then the shells are reassembledand bonded together with a belly bandof cured composite material. The re-sulting composite shell becomes a man-drel for the subsequent steps of the fab-

rication process and remains inside thefinal tank.

The outer surface of the compositeshell is covered with a layer of materialdesigned to be impermeable by the pres-surized fluid to be contained in the tank.A second step on the outer flange ofeach end fitting accommodates thislayer. Depending on the application, thislayer could be, for example, a layer ofrubber, a polymer film, or an electrode-posited layer of metal. If the fluid to becontained in the tank is a gas, then thebest permeation barrier is electrode-posited metal (typically copper ornickel), which can be effective at a thick-ness of as little as 0.005 in (≈0.13 mm).The electrodeposited metal becomes

molecularly bonded to the second stepon each metallic end fitting.

The permeation-barrier layer is cov-ered with many layers of filament-woundcomposite material, which could be thesame as, or different from, the compos-ite material of the inner shell. Finally,the filament-wound composite materialis cured in an oven.

This work was done by Tom DeLay of Mar-shall Space Flight Center.

This invention is owned by NASA, and apatent application has been filed. For further in-formation, contact Sammy Nabors, MSFC Com-mercialization Assistance Lead, at (256) 544-5226 or [email protected]. Refer toMFS-31594.

End Fitting

End Fitting

Permeation Barrier

Inner Half ShellsJoined Here

Thick Composite-MaterialOuter Shell

Thin Composite-MaterialInner Half Shells

Belly Band

This Lightweight Tank is a multilayer unitary structure. Each layer makes a distinct contribution to theoverall effectiveness of the structure.

Multilayer Composite Pressure VesselsLightweight tanks can be made in diverse sizes and shapes.Marshall Space Flight Center, Alabama


Bio-Medical

Texturing Blood-Glucose-Monitoring Optics Using Oxygen Beams Textures can be tailored to exclude blood cells from optical-sensing regions. John H. Glenn Research Center, Cleveland, Ohio

A method has been invented for utiliz-ing directed, hyperthermal oxygenatoms and ions for texturing tips of poly-meric optical fibers or other polymericoptical components for use in opticalmeasurement of concentration of glu-cose in blood. The required texture ofthe sensory surface of such a componentamounts to a landscape of microscopichills having high aspect ratios (hillstaller than they are wide), with an aver-age distance between hills of no morethan about 5 µm. This limit on the aver-age distance between hills is chosen sothat blood cells (which are wider) can-not enter the valleys between the hills,where they could obstruct optical sens-ing of glucose in the blood plasma. Onthe other hand, the plasma is requiredto enter the valleys, and a high aspectratio is intended to maximize the hill-side and valley surface area in contactwith the plasma, thereby making it possi-ble to obtain a given level of optical glu-cose-measurement sensitivity with a rela-tively small volume of blood.

The present method of texturing by useof directed, hyperthermal (particle energy>1 eV) oxygen atoms and ions stands incontrast to a prior method of texturing byuse of thermal monatomic oxygen charac-terized by a temperature of the order of0.5 eV. The prior method yields low-as-pect-ratio (approximately hemispherical)craters that are tens of microns wide —too wide to exclude blood cells.

The figure schematically depicts partsof a typical apparatus for texturing ac-cording to the present method. One ormore polymeric optical components to betextured (e.g., multiple optical fibers bun-

dled together for simultaneous process-ing) are mounted in a vacuum chamberfacing a suitable ion- or atom-acceleratingdevice capable of generating a beam ofoxygen atoms and/or ions having kineticenergies >1 eV. Typically, such a device in-cludes a heated cathode, in which case itis desirable to interpose a water-cooledthermal-radiation shield to prevent melt-ing of the polymeric component(s) to betextured. In operation, the chamber isevacuated to a pressure ≤10–5 torr (lessthan or equal to approximately 1.3 mPa),then the beam is turned on.

The resulting texture is characterized byapproximately conical hills having aspect

ratios greater than 1. In experiments, it wasdemonstrated that separations between ad-jacent hills can be made ≤1 µm and thatthe separations and heights of the hillscan be varied by varying the fluence ofmonatomic oxygen and/or oxygen ions.

This work was done by Bruce Banks of GlennResearch Center. Further information is con-tained in a TSP (see page 1).

Inquiries concerning rights for the commercialuse of this invention should be addressed toNASA Glenn Research Center, Commercial Tech-nology Office, Attn: Steve Fedor, Mail Stop 4–8,21000 Brookpark Road, Cleveland, Ohio44135. Refer to LEW-17642-1

+

–

DC PowerSupply

25 to 150 V

CathodeHeater

Power Supply

Water-CooledCathode

Heat-RadiationShield

Bundle of PolymericOptical Fibers

To Be Textured

O O+ O2+

O2

End-HallIon Source

Magnified Cross-Section ofTip of Textured Optical Fiber

HotCathode

Directed Energetic Oxygen Atoms and Ions impinging on tips of polymeric optical fibers cause the tipsto become textured with microscopic, approximately conical hills having large aspect ratios.


Physical Sciences

Fault-Tolerant Heat ExchangerA single-point leak would not cause mixing of heat-transfer fluids.Lyndon B. Johnson Space Center, Houston, Texas

A compact, lightweight heat ex-changer has been designed to be fault-tolerant in the sense that a single-pointleak would not cause mixing of heat-transfer fluids. This particular heat ex-changer is intended to be part of thetemperature-regulation system for hab-itable modules of the InternationalSpace Station and to function withwater and ammonia as the heat-transferfluids. The basic fault-tolerant design isadaptable to other heat-transfer fluidsand heat exchangers for applications inwhich mixing of heat-transfer fluidswould pose toxic, explosive, or otherhazards: Examples could includefuel/air heat exchangers for thermalmanagement on aircraft, process heatexchangers in the cryogenic industry,and heat exchangers used in chemicalprocessing.

The reason this heat exchanger cantolerate a single-point leak is that theheat-transfer fluids are everywhere sepa-rated by a vented volume and at least twoseals. The combination of fault toler-ance, compactness, and light weight isimplemented in a unique heat-ex-changer core configuration: Each fluidpassage is entirely surrounded by avented region bridged by solid struc-tures through which heat is conductedbetween the fluids. Precise, proprietaryfabrication techniques make it possibleto manufacture the vented regions and

heat-conducting structures with verysmall dimensions to obtain a very largecoefficient of heat transfer between thetwo fluids. A large heat-transfer coeffi-cient favors compact design by making itpossible to use a relatively small core fora given heat-transfer rate.

Calculations and experiments haveshown that in most respects, the fault-tol-erant heat exchanger can be expected toequal or exceed the performance of the

non-fault-tolerant heat exchanger that itis intended to supplant (see table). Theonly significant disadvantages are a slightweight penalty and a small decrease inthe mass-specific heat transfer.

This work was done by Michael G. Izensonand Christopher J. Crowley of Creare, Inc., forJohnson Space Center. For further infor-mation, contact the Johnson CommercialTechnology Office at (281) 483-3809.MSC-23271

Characteristic

Heat-TransferLoad

Mass (for Pinch-Point Criterion) <12 kg 14.6 kg

>1.2 kW/kg 0.96 kW/kg

>2 kW/kg 1.7 kW/kg

3.7 MPa 3.7 MPa

5.6 MPa 5.6 MPa

11.2 MPa 22.4 MPa

19 kPa 19 kPa

44.2 kPa 44.2 kPa

Volume and Dimensions toSatisfy Pinch-Point Criterion

Pressure Drop on Primary(H2O) Side at Mass Flow Rateof 380 g/s

6.55 L6.4 by 20.8 by 49.5 cm

2.5 L9.1 by 12.7 by 21.6 cm

14 kW 14 kW

25 kW 25 kW

Design Point

Pinch Point

Non-Fault-Tolerant Design Fault-Tolerant Design

Mass-SpecificHeat Transfer

Pressure

Design Point

Maximum Allow-able Working

Pinch Point

Proof

Design/Burst

Pressure Drop on Secondary(NH3) Side at Flow Rate of440 g/s

Design and Performance Characteristics of the fault-tolerant heat exchanger are shown alongsidethose of the prior non-fault-tolerant heat exchanger.

A proposed highly accurate clock or os-cillator would be based on the concept ofan opto-electronic oscillator (OEO) stabi-lized to an atomic transition. Opto-elec-tronic oscillators, which have been de-scribed in a number of prior NASA TechBriefs articles, generate signals at frequen-cies in the gigahertz range characterizedby high spectral purity but not by long-term stability or accuracy. On the other

hand, the signals generated by previouslydeveloped atomic clocks are character-ized by long-term stability and accuracybut not by spectral purity. The proposedatomic clock would provide high spectralpurity plus long-term stability and accu-racy — a combination of characteristicsneeded to realize advanced developmentsin communications and navigation. In ad-dition, it should be possible to miniaturize

the proposed atomic clock.When a laser beam is modulated by a

microwave signal and applied to a pho-todetector, the electrical output of thephotodetector includes a component atthe microwave frequency. In atomicclocks of a type known as Raman clocksor coherent-population-trapping (CPT)clocks, microwave outputs are obtainedfrom laser beams modulated, in each

Atomic Clock Based on Opto-Electronic OscillatorThis apparatus would afford spectral purity plus long-term stability and accuracy.NASA’s Jet Propulsion Laboratory, Pasadena, California


case, to create two sidebands that differin frequency by the amount of a hyper-fine transition in the ground state ofatoms of an element in vapor form in acell. The combination of these sidebandsproduces a transparency in the popula-tion of a higher electronic level that canbe reached from either of the twoground-state hyperfine levels by absorp-tion of a photon. The beam is transmit-ted through the vapor to a photodetec-tor. The components of light scattered ortransmitted by the atoms in the two hy-

perfine levels mix in the photodetectorand thereby give rise to a signal at the hy-perfine-transition frequency.

The proposed atomic clock would in-clude an OEO and a rubidium- or ce-sium-vapor cell operating in theCPT/Raman regime (see figure). In theOEO portion of this atomic clock, as ina typical prior OEO, a laser beam wouldpass through an electro-optical modula-tor, the modulated beam would be fedinto a fiber-optic delay line, and the de-layed beam would be fed to a photode-

tector. The electrical output of the pho-todetector would be detected, amplified,filtered, and fed back to the microwaveinput port of the modulator.

The laser would be chosen to have thesame wavelength as that of the pertinentground-state/higher-state transition ofthe atoms in the vapor. The modulator/filter combination would be designed tooperate at the microwave frequency ofthe hyperfine transition. Part of the laserbeam would be tapped from the fiber-optic loop of the OEO and introducedinto the vapor cell. After passing throughthe cell, this portion of the beam wouldbe detected differentially with a tappedportion of the fiber-optically-delayedbeam. The electrical output of the pho-todetector would be amplified and fil-tered in a loop that would control a DCbias applied to the modulator. In thismanner, the long-term stability and accu-racy of the atomic transition would betransferred to the OEO.

This work was done by Lute Maleki and NanYu of Caltech for NASA’s Jet Propulsion Lab-oratory. Further information is contained in aTSP (see page 1).

In accordance with Public Law 96-517,the contractor has elected to retain title to thisinvention. Inquiries concerning rights for itscommercial use should be addressed to:

Innovative Technology Assets ManagementJPLMail Stop 202-2334800 Oak Grove DrivePasadena, CA 91109-8099(818) 354-2240E-mail: [email protected] to NPO-30557, volume and number

of this NASA Tech Briefs issue, and thepage number.

Laser Modulator

Optical Delay (Fiber orOther Optical High Q Device)

Detector

BeamSplitter

Phase Shifter

RF Filter

Integrator

VaporCell

Raman AbsorptionError Signal Generation

Amplifier

Amplifier

This Atomic Clock would incorporate a conventional atomic clock and an opto-electronic oscillatorin such a manner as to exploit the best features of both.

Microfocus/Polycapillary-Optic Crystallographic X-Ray SystemThis system generates an intense, nearly collimated beam suitable for crystallography.Marshall Space Flight Center, Alabama

A system that generates an intense,nearly collimated, nearly monochro-matic, small-diameter x-ray beam hasbeen developed for use in macromolecu-lar crystallography. A conventional x-raysystem for macromolecular crystallogra-phy includes a rotating-anode x-raysource, which is massive (≥500 kg), large(approximately 2 by 2 by 1 m), andpower-hungry (between 2 and 18 kW). Incontrast, the present system generates abeam of the required brightness from amicrofocus source, which is small and

light enough to be mounted on a labora-tory bench, and operates at a power levelof only tens of watts.

The figure schematically depicts thesystem as configured for observing x-raydiffraction from a macromolecular crys-tal. In addition to the microfocus x-raysource, the system includes a polycapil-lary optic — a monolithic block (typi-cally a bundle of fused glass tubes) thatcontains thousands of straight or gentlycurved capillary channels, along whichx-rays propagate with multiple reflec-

tions. This particular polycapillary opticis configured to act as a collimator; thex-ray beam that emerges from its outputface consists of quasi-parallel subbeamswith a small angular divergence and adiameter comparable to the size of acrystal to be studied. The gap betweenthe microfocus x-ray source and theinput face of the polycapillary optic ischosen consistently with the focallength of the polycapillary optic and theneed to maximize the solid angle sub-tended by the optic in order to maxi-

mize the collimated x-ray flux. The spec-trum from the source contains a signifi-cant component of Cu Kα (photon en-ergy is 8.08 keV) radiation. The beam ismonochromatized (for Cu Kα) by anickel filter 10 μm thick.

In a test, this system was operated at apower of 40 W (current of 897 μA at anaccelerating potential of 45 kV), with ananode x-ray spot size of 41±2 μm. Alsotested, in order to provide a standard forcomparison, was a commercial rotating-

anode x-ray crystallographic system witha pyrolytic graphite monochromator anda 250-μm pinhole collimator, operatingat a power of 3.15 kW (current of 70 mAat an accelerating potential of 45 kV).The flux of collimated Cu Kα radiationin this system was found to be approxi-mately 16 times that in the rotating-anode system. Data on x-ray diffractionfrom crystals of tetragonal form oflysozyme (protein) in this system werefound to be of high quality and to be re-

ducible by use of standard crystallo-graphic software.

This work was done by Marshall Joy ofMarshall Space Flight Center, MikhailGubarev of the National Research Council,and Ewa Ciszak of Universities Space Re-search Association.

This invention is owned by NASA, and apatent application has been filed. For furtherinformation, contact Jim Dowdy, MSFCCommercialization Assistance Lead, [email protected]. Refer to MFS-31499.


This Microfocus-Source/Polycapillary-Collimator X-Ray System generates an x-ray flux greater than that of a larger and more power-hungry rotating-anode x-ray system.

MicrofocusX-Ray Source Polycapillary Collimator

Nickel Filter

0.5 mm2.6mm

10 mm MacromolecularCrystal

Diffraction Pattern on X-Ray Detector

Depth-Penetrating Luminescence Thermography of Thermal-Barrier Coatings Temperatures at depths can be measured by luminescence decay of suitable thermographicphosphors.John H. Glenn Research Center, Cleveland, Ohio

A thermographic method has been de-veloped for measuring temperatures at pre-determined depths within dielectric mate-rial layers ⎯ especially thermal-barriercoatings. This method will help satisfy aneed for noncontact measurement ofthrough-the-thickness temperature gradi-ents for evaluating the effectiveness of ther-mal-barrier coatings designed to preventoverheating of turbine blades, combustorliners, and other engine parts.

Heretofore, thermography has beenlimited to measurement of surface andnear-surface temperatures. In the thermo-graphic method that is the immediate

predecessor of the present method, a ther-mographic phosphor is applied to theouter surface of a thermal barrier coating,luminescence in the phosphor is excitedby illuminating the phosphor at a suitablewavelength, and either the time depend-ence of the intensity of luminescence orthe intensities of luminescence spectrallines is measured. Then an emissivity-inde-pendent surface-temperature value is com-puted by use of either the known tempera-ture dependence of the luminescencedecay time or the known temperature de-pendence of ratios between intensities ofselected luminescence spectral lines. Until

now, depth-penetrating measurementshave not been possible because light of thewavelengths needed to excite phosphorscould not penetrate thermal-barrier coat-ing materials to useful depths.

In the present method as in the methoddescribed above, one exploits the tempera-ture dependence of luminescence decaytime. In this case, the phosphor is incorpo-rated into the thermal-barrier coat at thedepth at which temperature is to be meas-ured. To be suitable for use in this method,a phosphor must (1) exhibit a temperaturedependence of luminescence decay timein the desired range, (2) be thermochemi-


cally compatible with the thermal-barriercoating, and (3) exhibit at least a minor ex-citation spectral peak and an emissionspectral peak, both peaks being at wave-lengths at which the thermal-barrier coat-ing is transparent or at least translucent.

Conventional thermographic phos-phors are not suitable because they aremost efficiently excited by ultravioletlight, which does not penetrate ther-mal-barrier coating materials. (Typicalthermal-barrier coating materials in-clude or consist of various formulationsof yttria-stabilized zirconia.) Only a

small fraction of phosphor candidateshave significant excitation at wave-lengths long enough (>500 nm) for suf-ficient penetration of thermal-barriercoatings. One suitable phosphor mate-rial ⎯ yttria doped with europium(Y2O3:Eu) ⎯ has a minor excitationpeak at 532 nm and an emission peak at611 nm. In experiments, this materialwas incorporated beneath a 100-μm-thick thermal-barrier coating and sub-jected to excitation and measurementby the luminescence-decay-time tech-nique. These experiments were found

to yield reliable temperature values upto 1,100 °C. At the time of reporting theinformation for this article, a search forsuitable phosphors other than(Y2O3:Eu) was continuing.

This work was done by Jeffrey Eldridge ofGlenn Research Center. Further informa-tion is contained in a TSP (see page 1).

Inquiries concerning rights for the commer-cial use of this invention should be addressed toNASA Glenn Research Center, CommercialTechnology Office, Attn: Steve Fedor, Mail Stop4-8, 21000 Brookpark Road, Cleveland Ohio44135. Refer to LEW-17617-1.

One-Dimensional Photonic Crystal SuperprismsIn comparison with three-dimensional superprisms, these could be fabricated more easily.NASA’s Jet Propulsion Laboratory, Pasadena, California

Theoretical calculations indicate thatit should be possible for one-dimen-sional (1D) photonic crystals (see fig-ure) to exhibit giant dispersions knownas the superprism effect. Previously,three-dimensional (3D) photonic crys-tal superprisms have demonstratedstrong wavelength dispersion — about500 times that of conventional prismsand diffraction gratings. Unlike diffrac-tion gratings, superprisms do not ex-hibit zero-order transmission or higher-order diffraction, thereby eliminatingcross-talk problems. However, the fabri-cation of these 3D photonic crystals re-quires complex electron-beam substratepatterning and multilayer thin-filmsputtering processes.

The proposed 1D superprism is muchsimpler in structural complexity and,therefore, easier to design and fabri-cate. Like their 3D counterparts, the 1Dsuperprisms can exhibit giant disper-sions over small spectral bands that canbe tailored by judicious structure designand tuned by varying incident beam di-rection. Potential applications includeminiature gas-sensing devices.

This work was done by David Ting of Cal-tech for NASA’s Jet Propulsion Labora-tory. Further information is contained in aTSP (see page 1).

This invention is owned by NASA, and apatent application has been filed. Inquiries

concerning nonexclusive or exclusive licensefor its commercial development should be ad-dressed to the Patent Counsel, NASA Man-agement Office– JPL at (818) 354-7770.Refer to NPO-30232.

A One-Dimensional Superprism could be fabricated more easily than a three-dimensional superprism. Likea three-dimensional prism, it would exhibit strong wavelength dispersion: two beams of light incident atthe same angle and having slightly different wavelengths would be refracted at two widely different angles.

Beam ofWavelength 2

Beam ofWavelength 1

Superprism


Information Sciences

Measuring Low-Order Aberrations in a Segmented TelescopeThis algorithm requires less computation than prescription-retrieval algorithms do.NASA’s Jet Propulsion Laboratory, Pasadena, California

The in-focus PSF optimizer (IPO) isan algorithm for use in monitoring andcontrolling the alignment of the seg-ments of a segmented-mirror astronom-ical telescope. IPO is so named becauseit computes wave-front aberrations ofthe telescope from digitized point-spread functions (PSFs) measured in in-focus images. Inasmuch as distant astro-nomical objects that behave optically aspoint sources can typically be seen in al-most any astronomical image, the mainbenefit afforded by IPO may be to en-able maintenance of mirror-segmentalignments without detracting fromvaluable scientific-observation time.

IPO evolved from prescription-re-trieval type algorithms. Prescription re-trieval uses in-focus and out-of-focusPSFs to infer the state of an imaging op-tical system. The state, in this context,

refers to the positions, orientations,and low-order figure errors of the opti-cal elements in the system. Both pre-scription-retrieval and IPO use an itera-tive, nonlinear, least-squares optimizerto compute the optimal state parame-ters such that a digital computer-gener-ated model image matches the digi-tized image acquired from the realsystem.

The difference between IPO and pre-scription-retrieval algorithms is thatIPO is specifically designed to utilize in-focus images only. Although the restric-tion to in-focus images limits IPO to cal-culating only the lowest-order wavefront aberrations, it also causes the re-sulting computation to take much lesstime because fewer degrees of freedomare included in the optimizationprocess.

In the prescription retrieval software de-veloped at JPL, the model images are gen-erated using the ray-trace/physical opticsprogram, MACOS. IPO, on the otherhand, uses a linear sensitivity matrix tocompute the exit-pupil wave front fromthe system parameters; the wave front isthen converted into a complex pupil field,which is then propagated to the imageplane via a fast Fourier transform. This ap-proach is computationally faster and re-quires less computer memory than isneeded for prescription retrieval.

This work was done by Catherine Ohara,David Redding, Fang Shi, Joseph Green, PhilipDumont, Scott Basinger, and Andrew Lowmanof Caltech for NASA’s Jet Propulsion Labora-tory and Laura Burns and Peter Petrone ofGoddard Space Flight Center. Further informa-tion is contained in a TSP (see page 1). NPO-30733

An algorithm has been developed tosolve the problem of mapping from(1) a glove instrumented with joint-angle sensors to (2) an anthropomor-phic robot hand. Such a mapping isneeded to generate control signals tomake the robot hand mimic the con-figuration of the hand of a human at-tempting to control the robot. Themapping problem is complicated byuncertainties in sensor locationscaused by variations in sizes andshapes of hands and variations in thefit of the glove. The present mappingalgorithm is robust in the face of theseuncertainties, largely because it in-cludes a calibration subalgorithm thatinherently adapts the mapping to thespecific hand and glove, without needfor measuring the hand and withoutregard for goodness of fit.

The algorithm utilizes a forward-kinematics model of the glove derivedfrom documentation provided by the

manufacturer of the glove. In this case,“forward-kinematics model” signifies amathematical model of the glove fin-gertip positions as functions of the sen-sor readings. More specifically, giventhe sensor readings, the forward-kine-matics model calculates the glove fin-gertip positions in a Cartesian refer-ence frame nominally attached to thepalm.

The algorithm also utilizes an inverse-kinematics model of the robot hand. Inthis case, “inverse-kinematics model”signifies a mathematical model of therobot finger-joint angles as functions ofthe robot fingertip positions. Again,more specifically, the inverse-kinematicsmodel calculates the finger-joint com-mands needed to place the fingertips atspecified positions in a Cartesian refer-ence frame that is attached to the palmof the robot hand and that nominallycorresponds to the Cartesian referenceframe attached to the palm of the glove.

Initially, because of the aforemen-tioned uncertainties, the glove finger-tip positions calculated by the forward-kinematics model in the gloveCartesian reference frame cannot beexpected to match the robot fingertippositions in the robot-hand Cartesianreference frame. A calibration must beperformed to make the glove androbot-hand fingertip positions corre-spond more precisely. The calibrationprocedure involves a few simple handposes designed to provide well-definedfingertip positions. One of the poses isa fist. In each of the other poses, a fin-ger touches the thumb. The calibra-tion subalgorithm uses the sensorreadings from these poses to modifythe kinematical models to make thetwo sets of fingertip positions agreemore closely.

In tests of software that implementsthe algorithm, the entire calibrationprocess was found to take less than 30

Mapping From an Instrumented Glove to a Robot HandFingertip positions can be made to match in a fast, simple calibration procedure.Lyndon B. Johnson Space Center, Houston, Texas


seconds. Operators immediately noted adifference between the accuracy of fin-gertip positions as computed by this algo-rithm and as computed by a prior algo-rithm. The increased accuracy afforded

by this algorithm was found to improvecontrol of the robot hand. The algo-rithm and software were also adapted touse with an optically tracked glove forhand control, with similar results.

This work was done by Michael Goza ofJohnson Space Center. For further infor-mation, contact the Johnson CommercialTechnology Office at (281) 483-3809. MSC-23680


Application of the Hilbert-Huang Transform to Financial Data

A paper discusses the application ofthe Hilbert-Huang transform (HHT)method to time-series financial-marketdata. The method was described, vari-ously without and with the HHT name,in several prior NASA Tech Briefs arti-cles and supporting documents. To re-capitulate: The method is especiallysuitable for analyzing time-series datathat represent nonstationary and non-linear phenomena including physicalphenomena and, in the present case,financial-market processes. Themethod involves the empirical modedecomposition (EMD), in which acomplicated signal is decomposed intoa finite number of functions, called“intrinsic mode functions” (IMFs),that admit well-behaved Hilbert trans-forms. The HHT consists of the combi-nation of EMD and Hilbert spectralanalysis. The local energies and the in-stantaneous frequencies derived fromthe IMFs through Hilbert transformscan be used to construct an energy-fre-quency-time distribution, denoted aHilbert spectrum. The instant paperbegins with a discussion of prior ap-proaches to quantification of marketvolatility, summarizes the HHTmethod, then describes the applica-tion of the method in performingtime-frequency analysis of mortgage-market data from the years 1972through 2000. Filtering by use of theEMD is shown to be useful for quanti-fying market volatility.

This work was done by Norden Huang ofGoddard Space Flight Center. Further in-formation is contained in a TSP (see page 1).

This invention is owned by NASA, and apatent application has been filed. Inquiriesconcerning nonexclusive or exclusive licensefor its commercial development should be ad-dressed to the Patent Counsel, Goddard SpaceFlight Center, (301) 286-7351. Refer to GSC-14807-1.

Optimizing Parameters for Deep-Space OpticalCommunication

A paper discusses the optimization ofthe parameters of a high-rate, deep-spaceoptical communication link that utilizespulse-position modulation (PPM) and anerror-correcting code (ECC). The pa-rameters in question include the PPMorder (number of pulse time slots in onesymbol period), the ECC rate, and theuncoded symbol error rate. In simpleterms, the optimization problem is tochoose the combination of these param-eters that maximizes the throughputdata rate at a given bit-error-rate (BER),subject to several constraints, includinglimits on the average and peak powerand possibly a limit on the uncoded sym-bol error rate. This is a complex, multidi-mensional optimization problem, the so-lution of which involves computation ofchannel capacities for various combina-tions of the parameters. The paper pres-ents extensive theoretical analyses andnumerical predictions that elucidate themany facets of the optimization problem.It shows how a nearly optimum solution

can be obtained by choosing the opti-mum PPM order for the desired numberof bits per slot and concatenating thePPM mapping with an error-correctioncode so that the decoded bits satisfysome BER threshold.

This work was done by Bruce Moison andJon Hamkins of Caltech for NASA’s JetPropulsion Laboratory. Further informa-tion is contained in a TSP (see page 1).NPO-40591

Low-Shear Microencapsulation andElectrostatic Coating

A report presents additional informa-tion on the topic of a microencapsulationelectrostatic processing system. Informa-tion in the report includes micrographsof some microcapsules, a set of diagramsthat schematically depict the steps of anencapsulation process, and brief descrip-tions of (1) alternative versions of thepres- ent encapsulation processes, (2) ad-vantages of the present microencapsula-tion processes over prior microencapsula-tion processes, and (3) unique andadvantageous features of microcapsulesproduced by the present processes.

This work was done by Dennis R. Morrisonof Johnson Space Center and BenjaminMosier of the Institute for Research, Inc.

This invention has been patented by NASA(U.S. Patent No. 6,103,271). Inquiries con-cerning nonexclusive or exclusive license forits commercial development should be ad-dressed to the Patent Counsel, Johnson SpaceCenter, (281) 483-0837. Refer to MSC-22938.

Books & Reports

National Aeronautics andSpace Administration

technology focus computers/electronics - nasa immersed in the liquid. ... any gas bubble flowing ......

Documents