technology focus computers/electronics

Technology Focus

Computers/Electronics

Software

Materials

Mechanics

Machinery/Automation

Manufacturing

Bio-Medical

Physical Sciences

Information Sciences

Books and Reports

10-05 October 2005

NASA Tech Briefs, October 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 Program (IPP), its documents, and services is also available at the same facility oron the World Wide Web at http://ipp.nasa.gov.

Innovative Partnerships Offices are located at NASA field centers to provide technology-transfer access toindustrial users. Inquiries can be made by contacting NASA field centers and Mission Directorates 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 CenterHelen Lane(713) [email protected]

Kennedy Space CenterJim Aliberti(321) [email protected]

Langley Research CenterRay P. Turcotte(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 Innovation Research Program (SBIR) &Small Business TechnologyTransfer Program (STTR)(202) [email protected]

Frank SchowengerdtInnovative Partnerships Program(Code TD)(202) [email protected]

John MankinsExploration Systems Researchand Technology Division(202) [email protected]

Terry HertzAeronautics and Space MissionDirectorate(202) [email protected]

Glen MucklowMission and Systems Management Division (SMD)(202) [email protected]

Granville PaulesMission and Systems Management Division (SMD)(202) [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 MMiissssiioonn DDiirreeccttoorraatteess

At NASA Headquarters there are four Mission Directorates underwhich there are seven major program offices that develop andoversee technology projects of potential interest to industry:

5 Technology Focus: Sensors5 Insect-Inspired Optical-Flow Navigation Sensors

6 Chemical Sensors Based on Optical RingResonators

7 A Broad-Band Phase-Contrast Wave-Front Sensor

8 Progress in Insect-Inspired Optical NavigationSensors

11 Electronics/Computers11 Portable Airborne Laser System Measures Forest-

Canopy Height

11 Deployable Wide-Aperture Array Antennas

12 Faster Evolution of More Multifunctional LogicCircuits

13 Video-Camera-Based Position-Measuring System

14 N-Type δ Doping of High-Purity Silicon ImagingArrays

15 Software15 Avionics System Architecture Tool

15 Updated Chemical Kinetics and SensitivityAnalysis Code

15 Predicting Flutter and Forced Response inTurbomachinery

15 Upgrades of Two Computer Codes for Analysis ofTurbomachinery

16 Program Facilitates CMMI Appraisals

16 Grid Visualization Tool

16 Program Computes Sound Pressures at RocketLaunches

16 Solar-System Ephemeris Toolbox

16 Data-Acquisition Software for PSP/TSP Wind-Tunnel Cameras

19 Materials19 Corrosion-Prevention Capabilities of a Water-

Borne, Silicone-Based, Primerless Coating

19 Sol-Gel Process for Making Pt-Ru Fuel-CellCatalysts

20 Making Activated Carbon for Storing Gas

21 Machinery/Automation21 System Regulates the Water Contents of Fuel-Cell

Streams

21 Five-Axis, Three-Magnetic-Bearing Dynamic SpinRig

23 Manufacturing23 Modifications of Fabrication of Vibratory

Microgyroscopes

25 Bio-Medical25 Chamber for Growing and Observing Fungi

25 Electroporation System for Sterilizing Water

27 Physical Sciences27 Thermoelectric Air/Soil Energy-Harvesting Device

27 Flexible Metal-Fabric Radiators

28 Actuated Hybrid Mirror Telescope

29 Optical Design of an Optical CommunicationsTerminal

31 Information Sciences31 Algorithm for Identifying Erroneous Rain-Gauge

Readings

31 Condition Assessment and End-of-Life PredictionSystem for Electric Machines and Their Loads

33 Books & Reports33 Lightweight Thermal Insulation for a Liquid-

Oxygen Tank

33 Stellar Gyroscope for Determining Attitude of aSpacecraft

33 Lifting Mechanism for the Mars Explorer Rover

10-05 October 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

Insect-Inspired Optical-Flow Navigation SensorsOptical mouse chips are used to measure optical flow.NASA’s Jet Propulsion Laboratory, Pasadena, California

Integrated circuits that exploit opticalflow to sense motions of computer miceon or near surfaces (“optical mousechips”) are used as navigation sensors ina class of small flying robots now under-going development for potential use insuch applications as exploration, search,and surveillance. The basic principles ofthese robots were described briefly in“Insect-Inspired Flight Control for SmallFlying Robots” (NPO-30545), NASA TechBriefs, Vol. 29, No. 1 (January2005), page 61. To recapitulatefrom the cited prior article: Theconcept of optical flow can be de-fined, loosely, as the use of tex-ture in images as a source of mo-tion cues. The flight-control andnavigation systems of these robotsare inspired largely by the designsand functions of the vision sys-tems and brains of insects, whichhave been demonstrated to uti-lize optical flow (as detected bytheir eyes and brains) resultingfrom their own motions in the en-vironment.

Optical flow has been shown tobe very effective as a means ofavoiding obstacles and control-ling speeds and altitudes in ro-botic navigation. Prior systemsused in experiments on navigat-ing by means of optical flow haveinvolved the use of panoramic op-tics, high-resolution image sen-sors, and programmable image-data-processing computers. Thesesystems are large, complex, and compu-tationally expensive, and not readily scal-able for inclusion in miniature robots,for which there are severe design re-quirements to limit power demand,mass, and size.

The present development exploits therecent proliferation and commercialavailability of optical mouse chips. Eachoptical mouse chip includes a low-resolu-tion (16 × 16) array of photosensors, andcircuitry that compares consecutiveimage frames to compute the optical flowacross the array in two dimensions, in amanner analogous to that of an elementin an insect’s compound eye. In a com-

puter mouse, the optical flow is used totrack the movement of the mouse on amouse pad or equivalent surface; in a fly-ing robot of the type now under develop-ment, the optical flow serves as a measureof two-dimensional velocity relative tonearby surfaces and objects. The use ofoptical mouse chips instead of the imag-ing-and-computing systems describedabove offers advantages of compactness,low mass (15 to 20 g per chip), low power

demand (42 mW per chip), low cost(about $10 per chip in year 2004), redun-dancy, high speed (frame rates up to 2.3kHz), and parallel processing.

In a hierarchical control architectureproposed for subsequent development ofa flying robot, the outputs of several opti-cal-mouse-type navigation sensors wouldbe fed to a microcontroller (see figure)that would utilize the combined optical-flow information to determine the mo-tion of the robot relative to the environ-ment. This microcontroller would, inturn, communicate with a master micro-controller, which would combine infor-mation from various sensing subsystems,determine the priority to be assigned to

the information from each subsystem,and relay control information to affect lo-comotion. This hierarchical architectureis analogous to the neural structures offlies.

This work was done by Sarita Thakoorand John M. Morookian of Caltech; JavanChahl and Dean Soccol of Australian Na-tional University; and Butler Hine andSteven Zornetzer of NASA Ames ResearchCenter for NASA’s Jet Propulsion Labora-

tory. 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-40173, volume and number

of this NASA Tech Briefs issue, and thepage number.

Outputs of Optical Mouse Chips would be fed to an optical-flow microcontroller for use in controlling a smallflying robot. The number of mouse chips needed would increase with the required complexity of the behav-ior of the robot.

OpticalMouseChip

Lens

OpticalMouseChip

Lens

OpticalMouseChip

Lens

OpticalMouseChip

Lens

Optical-FlowMicrocontroller

To MasterMicrocontroller

Serial Data

Serial Data

Serial Data

Serial Data

Serial ClockSignal

6 NASA Tech Briefs, October 2005

Chemical Sensors Based on Optical Ring ResonatorsResonance wavelengths are shifted by absorption of chemicals into polymer cladding layers.NASA’s Jet Propulsion Laboratory, Pasadena, California

Chemical sensors based on optical ringresonators are undergoing development. Aring resonator according to this concept isa closed-circuit dielectric optical waveguide.The outermost layer of this waveguide,analogous to the optical cladding layer onan optical fiber, is a made of a polymer that(1) has an index of refraction lower thanthat of the waveguide core and (2) absorbschemicals from the surrounding air. Theindex of refraction of the polymer changeswith the concentration of absorbed chemi-cal(s). The resonator is designed to operatewith relatively strong evanescent-wave cou-pling between the outer polymer layer andthe electromagnetic field propagatingalong the waveguide core. By virtue of thiscoupling, the chemically induced changein index of refraction of the polymer causesa measurable shift in the resonance peaksof the ring.

In a prototype that has been used todemonstrate the feasibility of this sensorconcept, the ring resonator is a dielectricoptical waveguide laid out along a closedpath resembling a racetrack (see Figure1). The prototype was fabricated on a sili-con substrate by use of standard tech-niques of thermal oxidation, chemicalvapor deposition, photolithography, etch-ing, and spin coating. The prototype res-onator waveguide features an innercladding of SiO2, a core of SixNy, and achemical-sensing outer cladding of ethylcellulose. In addition to the ring res-onator, there are input and output wave-guides separated from the straight seg-ments of the ring resonator by anevanescent-wave-coupling gap of 2 mm.

Figure 2 presents results of a test of theprototype in an open room. During thetest, the temperature of the sensor was sta-bilized to ±0.1 K. The sensor was leftundisturbed by chemicals, except duringa short interval when a cotton swab wettedwith isopropyl was placed 4 in. (≈10 cm)away from the sensor and another shortinterval when a cotton swab wetted withacetone was similarly placed near the sen-sor. The chemical exposures resulted ineasily detectable signals that exceededbackground variations by at least an orderof magnitude. The jagged nature of theportions of the plot corresponding to thechemical exposures has been attributedto “mode hops,” in which the specificring-resonator mode that was being fol-lowed moved out of the tuning range of a

laser used as the input light source, caus-ing the laser to lock onto a new mode.

The results have been interpreted asdemonstrating the feasibility of opticalpolymer-based sensors. Inasmuch as theindex of refraction of ethyl cellulose isknown to respond to wide variety ofvolatiles, sensors like this one could beuseful as non-specific indicators of spillsof volatile compounds.

This work was done by Margie Homer, Alli-son Manfreda, Kamjou Mansour, Ying Lin,and Alexander Ksendzov of Caltech for NASA’sJet Propulsion Laboratory. Further informa-tion is contained in a TSP (see page 1).

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


of this NASA Tech Briefs issue, and the pagenumber.

Figure 1. A Polymer-Clad Optical Ring Resonator acts as a chemical sensor in that the resonance spec-trum becomes shifted in wavelength when the polymer absorbs chemicals from the air.

1,100 Å900 Å

0.2mm Radius =

1.5 mm

Input

ENLARGED CROSS SECTION OFRING RESONATOR WAVEGUIDE PLAN VIEW OF RING

RESONATOR AND INPUT ANDOUTPUT WAVEGUIDES

Output B

Output A

Notes:1. Not to scale.2. n denotes index of refraction.

2 μm

7 μm

1 μm

SixNy Core,n = 1.855 ± 0.004

Polymer (Ethyl Cellulose)Outer Cladding,n = 1.47 ± 0.02

Si SubstrateSi SubstrateSi Substrate

SiO2 Inner Cladding,n = 1.454 ± 0.004

Figure 2. Shifts in the Wavelength of a peak in the resonance spectrum of the device of Figure 1 oc-curred during exposure to chemicals deliberately introduced into the air.

Exposure toIsopropyl Alcohol

Exposure toAcetone

-0.05

-0.10

-0.15

-0.20

-0.25

-0.30

50 75 100

Elapsed Time, min

Shif

t in

Pea

k Po

siti

on

125 150


A broadband phase-contrast wave-frontsensor has been proposed as a real-timewave-front sensor in an adaptive-opticssystem. The proposed sensor would offeran alternative to the Shack-Hartmannwave-front sensors now used in high-order adaptive-optics systems of some as-tronomical telescopes. Broadband sens-ing gives higher sensitivity than doesnarrow-band sensing, and it appears thatfor a given bandwidth, the sensitivity ofthe proposed phase-contrast sensor couldexceed that of a Shack-Hartmann sensor.Relative to a Shack-Hartmann sensor, theproposed sensor may be optically andmechanically simpler.

As described below, an important ele-ment of the principle of operation of aphase-contrast wave-front sensor is the im-position of a 90° phase shift between dif-fracted and undiffracted parts of the samelight beam. In the proposed sensor, thisphase shift would be obtained by utilizingthe intrinsic 90° phase shift between thetransmitted and reflected beams in anideal (thin, symmetric) beam splitter. Thisphase shift can be characterized as achro-matic or broadband because it is 90° atevery wavelength over a broad wavelengthrange.

The phase-contrast approach was orig-inally devised by Frits Zernike for mi-croscopy as a means of obtaining inten-sity images from such phase objects astransparent biological samples. Figure 1schematically illustrates an adaptation ofthe phase-contrast approach to real-timewave-front sensing for adaptive optics.The incident light from a guide star canbe described in terms of a pupil fieldfunction Aexp(iφ), where A is an aper-ture function that expresses the effect ofthe shape and size of the telescope pupiland φ is the difference between the ac-tual instantaneous phase and the nomi-nal (e.g., plane-wave) phase of the wavefront at a given position within thepupil. If the pupil were simply re-im-aged, the phase signal would not nor-mally be observable. To make the phasesignal observable, one reasons as follows:

Assuming a small-signal approxima-tion (φ << 1), the phase part of the pupilfield function could be approximated as1 + iφ. Hence, the phase-difference (dif-fracted) component would be 90° out ofphase with the larger undiffracted com-ponent. To a first approximation, the un-

diffracted rays would be localized withinthe central ≈λ/D portion of the tele-scope focal plane (where λ is wavelengthand D is the diameter of the primary mir-ror or lens of the telescope), while thediffracted rays that contribute to thephase component would impinge on thetelescope focal plane at off-axis positions.If a ±90°-phase-shift filter (e.g., a dielec-

tric disk of suitable thickness) having ap-proximately the diffraction-limited sizeλ/D were placed at the focal point, thenthe undiffracted component would beshifted by ±90° and would thereby bebrought into phase (or phase opposi-tion) with the diffracted component. Asa result, the phase component would be-come observable as a small variation in

Figure 1. Phase-Contrast Imaging would use 90° phase shifting to generate phase feedback for adap-tive optics.

Light FromGuide Star D

Wave Front WithLocal Phase Shift φ

90° Phase-Shift Filter

FocalPlane

TelescopePupil Subapertures

d

Focal Spots WithPhotodetectors inSubapertures

A Broad-Band Phase-Contrast Wave-Front SensorThe intrinsic 90° phase shift of an ideal beam splitter would be exploited.NASA’s Jet Propulsion Laboratory, Pasadena, California

Figure 2. A Broad-Band Phase-Contrast Wave-Front Sensor, shown here schematically, can be realizedby using the intrinsic properties of a beam splitter to give an achromatic 90°-phase-shifting element.

Beam FromTelescope

Phase-ContrastOutput Port 1

Phase-ContrastOutput Port 2

Thin, Symmetric50:50 Beam Splitter

Pinhole in M2

M1

M1


intensity across the pupil, superposed onthe bright, uniform illumination of theundiffracted component. In the small-signal approximation, the total intensityin the re-imaged pupil would be propor-tional to 1±2φ, the sign of 2φ dependingon whether the focal-spot filter advancesor retards the phase.

Figure 2 schematically illustrates an op-tical assembly, according to the proposal,for implementing the 90°-phase-shift fil-ter needed in a phase-contrast sensor likethat of Figure 1. An incident beam froma telescope would strike a 50:50 beamsplitter. The reflected and transmittedbeams would be recombined by anarrangement of mirrors, schematicallyrepresented by flats M1 in Figure 2; one

component is directed through a diffrac-tion-limited pinhole in two-sided mirrorM2. The pinhole would pass the central≈λ/D portions of the beams, while theM2 surfaces surrounding the pinholewould reflect the off-axis portions. Thetotal beam going to the output port oneach side of M2 would comprise the de-sired combination of central rays and 90°-shifted off-axis rays. The output beamscould be directed into telescope-pupil-re-imaging optics equipped with a charge-coupled-device (CCD) or similar quan-tum detector, as in Figure 1. Optionally,the phase-contrast images contained inboth beams could be combined opticallyor electronically to increase the signal-to-noise ratio.

This work was done by Eric Bloemhof and J.Kent Wallace of Caltech for NASA’s JetPropulsion Laboratory. Further informationis contained in a TSP (see page 1).

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


of this NASA Tech Briefs issue, and the pagenumber.

Progress in Insect-Inspired Optical Navigation SensorsSome details of implementation have become available.NASA’s Jet Propulsion Laboratory, Pasadena, California

Progress has been made in continuingefforts to develop optical flight-controland navigation sensors for miniature ro-botic aircraft. The designs of these sen-sors are inspired by the designs andfunctions of the vision systems andbrains of insects. Two types of sensors ofparticular interest are polarization com-passes and ocellar horizon sensors.

The basic principle of polarizationcompasses was described (but withoutusing the term “polarization compass”)in “Insect-Inspired Flight Control forSmall Flying Robots” (NPO-30545),NASA Tech Briefs, Vol. 29, No. 1 (January2005), page 61. To recapitulate: Bees usesky polarization patterns in ultraviolet(UV) light, caused by Rayleigh scatteringof sunlight by atmospheric gas mole-cules, as direction references relative tothe apparent position of the Sun. A ro-botic direction-finding technique basedon this concept would be more robust incomparison with a technique based onthe direction to the visible Sun becausethe UV polarization pattern is distrib-uted across the entire sky and, hence, isredundant and can be extrapolated froma small region of clear sky in an else-where cloudy sky that hides the Sun.

Three different implementations of apolarization compass are under consider-ation. Each implementation offers dis-tinct advantages and disadvantages rela-tive to the others:• In the lightest and least power-con-

sumptive implementation, the polariza-

tion in the sky is sampled in, typically,10 fields of view, each centered on a dif-ferent direction and having an angularwidth between 10° and 20°. An eight-bit microcontroller suffices to do all re-quired data processing. A productionversion of a sensor according to this im-plementation could be self-contained.One disadvantage of this implementa-tion, as determined in experimentsperformed thus far, is that bearing ac-curacy is characterized by an uncer-tainty of about 2°. Another disadvan-

tage is that this sensor cannot be usedfor imaging.

• In the second implementation, three dif-ferently oriented polarization filters areused to produce three subimages of thesky scene in separate focal-plane areas ofa complementary metal oxide/semicon-ductor (CMOS) video camera (see fig-ure). This implementation is amenable tosophisticated processing of polarization-image data and possible sub-degree accu-racy in determining the relative angularposition of the Sun. Unfortunately, for a

PolarizationFilters

PolarizationSubimages

FocalPlane

Aperture

BlueFilter

Three Differently Oriented Polarization Filters are used in projecting subimages on a CMOS image de-tector. In addition, a short-wavelength-pass (blue) filter contributes to image contrast because the po-larization signal is strongest in blue light.


production version, power consumptionand mass would be much greater than inthe first-mentioned implementation be-cause an embedded computer or digitalsignal processor would be necessary forprocessing video data. Design and fabrica-tion of the camera optics would present achallenge, inasmuch as the field of viewshould, ideally, be 150° wide. The chal-lenge is compounded by the need toavoid reflective optics, which would dis-rupt the polarization pattern.

• In the most elegant implementation,not yet realized, each pixel of a charge-coupled-device (CCD) camera wouldbe subdivided into three subpixels,each covered with a differently ori-

ented polarization filter. The resultingdevice would be small and lightweightand would demand little power, butmanufacturing would be complex. Thebasic principle of ocellar horizon sen-sors was also described in the citedprior article. These sensors are basedpartly on dragonfly ocelli — simpleeyes that exist in addition to the better-known compound eyes of insects andthat sense only light, dark, and motion.In dragonflies, the ocelli play an impor-tant role in stabilizing attitude with re-spect to dorsal light levels.An ocellar horizon sensor of the type

under development includes UV/greenpairs of photodiodes and utilizes drag-

onfly-inspired principles of color-oppo-nency processing. The reason forchoosing UV and green is that at thesewavelengths, spectral sensitivity of drag-onfly ocelli and the contrast betweenthe sky and ground are greatest: OnEarth, the contrast is greatest in thenear UV during the day and is greatestin green at twilight.

This work was done by Sarita Thakoor of Cal-tech, Javaan Chahl of Australian NationalUniversity, and Steve Zornetzer of NASA AmesResearch Center for NASA’s Jet PropulsionLaboratory. For further information, contactthe JPL Innovative Partnerships Office at (818)354-3821. NPO-41269


Deployable Wide-Aperture Array AntennasAntennas would be unrolled or unfolded to full size when and where needed. Lyndon B. Johnson Space Center, Houston, Texas

Inexpensive, lightweight array anten-nas on flexible substrates are under devel-opment to satisfy a need for large-aper-ture antennas that can be storedcompactly during transport and de-ployed to full size in the field. Conceivedfor use aboard spacecraft, antennas ofthis type also have potential terrestrialuses ⎯ most likely, as means to extendthe ranges of cellular telephones in ruralsettings.

Several simple deployment mecha-nisms are envisioned. One example isshown in the figure, where the deploy-ment mechanism, a springlike materialcontained in a sleeve around theperimeter of a flexible membrane, isbased on a common automobile windowshade. The array can be formed of an-tenna elements that are printed on smallsections of semi-flexible laminates, orpreferably, elements that are con-structed of conducting fabric. Likewise,

Electronics/Computers

FlexibleSheet of DielectricMaterial

AntennaElements

PowerDistribution

A Wide Array of Four Radiating Antenna Elements and their transmission line would be made fromflexible conductive materials on a flexible dielectric sheet. When not in use, the antenna could berolled into a compact cylinder in the manner of a window shade.

Portable Airborne Laser System Measures Forest-CanopyHeightThis system can be built, operated, and repaired at relatively low cost.Goddard Space Flight Center, Greenbelt, Maryland

The Portable Airborne Laser System(PALS) is a combination of laser rang-ing, video imaging, positioning, anddata-processing subsystems designed formeasuring the heights of forest canopiesalong linear transects from tens to thou-sands of kilometers long. Unlike priorlaser ranging systems designed to servethe same purpose, the PALS is not re-stricted to use aboard a single aircraft ofa specific type: the PALS fits into twolarge suitcases that can be carried to anyconvenient location, and the PALS canbe installed in almost any local aircraftfor hire, thereby making it possible tosample remote forests at relatively lowcost. The initial cost and the cost of re-pairing the PALS are also lower becausethe PALS hardware consists mostly ofcommercial off-the-shelf (COTS) units

that can easily be replaced in the field.The COTS units include a laser rang-

ing transceiver, a charge-coupled-devicecamera that images the laser-illuminatedtargets, a differential Global PositioningSystem (dGPS) receiver capable of oper-ation within the Wide Area Augmenta-tion System, a video titler, a video cas-sette recorder (VCR), and a laptopcomputer equipped with two serialports. The VCR and computer are pow-ered by batteries; the other units arepowered at 12 VDC from the 28-VDC air-craft power system via a low-pass filterand a voltage converter.

The dGPS receiver feeds location andtime data, at an update rate of 0.5 Hz, tothe video titler and the computer. Thelaser ranging transceiver, operating at asampling rate of 2 kHz, feeds its serial

range and amplitude data stream to thecomputer. The analog video signal fromthe CCD camera is fed into the video ti-tler wherein the signal is annotated withposition and time information. The titlerthen forwards the annotated signal tothe VCR for recording on 8-mm tapes.The dGPS and laser range and ampli-tude serial data streams are processed bysoftware that displays the laser trace andthe dGPS information as they are fedinto the computer, subsamples the laserrange and amplitude data, interleavesthe subsampled data with the dGPS in-formation, and records the resulting in-terleaved data stream.

This work was done by Ross Nelson ofGoddard Space Flight Center. Further in-formation is contained in a TSP (see page 1).GSC-14906-1


a distribution network connecting theelements can be created from conven-tional technologies such as lightweight,flexible coaxial cable and a surfacemount power divider, or preferably,from elements formed from conductivefabrics. Conventional technologies maybe stitched onto a supporting flexiblemembrane or contained within pocketsthat are stitched onto a flexible mem-brane. Components created from con-ductive fabrics may be attached by stitch-ing conductive strips to a nonconductivemembrane, embroidering conductivethreads into a nonconductive mem-brane, or weaving predetermined pat-terns directly into the membrane.

The deployable antenna may com-prise multiple types of antenna ele-ments. For example, thin profile an-tenna elements above a ground plane,both attached to the supporting flexiblemembrane, can be used to create a uni-directional boresight radiation pattern.Or, antenna elements without a groundplane, such as bow-tie dipoles, can be at-tached to the membrane to create a bidi-rectional array such as that shown in thefigure. For either type of antenna ele-ment, the dual configuration, i.e., ele-ments formed of slots in a conductivemembrane, can also be used. Finally,wide bandwidth antennas or arrays canbe formed in which the principal direc-

tion of radiation is in the plane of themembrane. For this embodiment, theset of elements on the membrane isarranged to form one or more travelingwave antennas. In this case, a noncon-ductive form of the perimeter springlikematerial is required to provide the de-ploying force.

This work was done by Patrick W. Fink,Justin A. Dobbins, Greg Y. Lin, Andrew Chu,and Robert C. Scully of Johnson SpaceCenter. This invention is owned by NASA,and a patent application has been filed. In-quiries concerning nonexclusive or exclusive li-cense for its commercial development should beaddressed to the Patent Counsel, Johnson SpaceCenter, (281) 483-0837. Refer to MSC-23436.

Faster Evolution of More Multifunctional Logic CircuitsEvolution is driven to find circuits that perform larger numbers of logic functions.NASA’s Jet Propulsion Laboratory, Pasadena, California

A modification in a method of auto-mated evolutionary synthesis of voltage-controlled multifunctional logic circuitsmakes it possible to synthesize more cir-cuits in less time. Prior to the modifica-tion, the computations for synthesizing afour-function logic circuit by thismethod took about 10 hours. Using themethod as modified, it is possible to syn-thesize a six-function circuit in less thanhalf an hour.

The concepts of automated evolution-ary synthesis and voltage-controlled multi-functional logic circuits were described ina number of prior NASA Tech Briefs articles.To recapitulate: A circuit is designed toperform one of several different logicfunctions, depending on the value of anapplied control voltage. The circuit designis synthesized following an automated evo-lutionary approach that is so named be-cause it is modeled partly after the repeti-tive trial-and-error process of biologicalevolution. In this process, random popula-tions of integer strings that encode elec-tronic circuits play a role analogous to thatof chromosomes. An evolved circuit istested by computational simulation (priorto testing in real hardware to verify a finaldesign). Then, in a fitness-evaluation step,responses of the circuit are compared withspecifications of target responses andcircuits are ranked according to howclose they come to satisfying speci-fications. The results of the evaluationprovide guidance for refining designsthrough further iteration.

As described in more detail in the prior

NASA Tech Briefs articles on multifunc-tional logic circuits, the multiple function-ality of these circuits, the use of a singlecontrol voltage to select the function, andthe automated evolutionary approach tosynthesis, offer potential advantages forthe further development of field-pro-grammable gate arrays (FPGAs):• Typical circuitry can be less complex

and can occupy smaller areas; becauseonly a single analog control line isneeded to select different functions.

• If voltage-controlled multifunctionalgates were used in the place of the con-figurable logic blocks of present com-mercial FPGAs, it would be possible tochange the functions of the resultingdigital systems in much shorter times;

• Relative to conventional circuits de-signed to perform single functions,multifunctional circuits can be synthe-

sized to be more tolerant of radiation-induced faults.In the unmodified method of auto-

mated evolutionary synthesis, the targetresponses of a multifunctional logic cir-cuit are fixed: that is, the user specifies inadvance which logic function the circuitis to perform at each of several discretevalues of control voltage (for example,AND at 0 V, NOR at 0.9 V, and NAND at1.8 V). In the modified method, the userno longer specifies which logic functionoccurs at which control voltage: Instead,the evolutionary algorithm is allowed tofind the control-voltage levels at whichvarious logic functions appear, and thefitness-evaluation function is modified toassign a higher fitness score to a circuitthat exhibits a greater number of logicfunctions over the full range of the con-trol voltage. Thus, evolution is driven to

Input 1

Buffer

Buffered(Digitized)

Output

Buffered Output

Direct Output

Time

Ou

tpu

t V

olt

age

Direct Output

EvolvingCircuitInput 2

ControlVoltage

A Buffer Must Be Added to an evolving circuit to obtain a digital output because the direct output ofa typical evolving circuit is an analog voltage.


Video-Camera-Based Position-Measuring SystemCoordinates of nearby targeted objects are measured quickly, easily, and safely.John F. Kennedy Space Center, Florida

A prototype optoelectronic systemmeasures the three-dimensional relativecoordinates of objects of interest or of tar-gets affixed to objects of interest in aworkspace. The system includes acharge-coupled-device video cameramounted in a known position and orien-tation in the workspace, a frame grabber,and a personal computer running image-data-processing software. Relative to con-ventional optical surveying equipment,this system can be built and operated atmuch lower cost; however, it is less accu-rate. It is also much easier to operate thanare conventional instrumentation sys-tems. In addition, there is no need to es-tablish a coordinate system through co-operative action by a team of surveyors.

The system operates in real time ataround 30 frames per second (limitedmostly by the frame rate of the camera).It continuously tracks targets as long asthey remain in the field of the camera.In this respect, it emulates more expen-sive, elaborate laser tracking equipmentthat costs of the order of 100 times asmuch. Unlike laser tracking equipment,this system does not pose a hazard oflaser exposure.

Images acquired by the camera aredigitized and processed to extract allvalid targets in the field of view. Thethree-dimensional coordinates (x, y, andz) of each target are computed from thepixel coordinates of the targets in theimages to accuracy of the order of mil-

limeters over distances of the orders ofmeters. The system was originally in-tended specifically for real-time positionmeasurement of payload transfers frompayload canisters into the payload bay of

Figure 1. In this Laboratory Setup, the camera of the prototype system is aimed at a mockup of a latchmating with a trunnion to demonstrate the use of the system to measure the three-dimensional co-ordinates of the latch relative of the trunnion.

Figure 2. A Target Pattern of Light and DarkSquares is processed by a block convolutionmask to obtain a pattern of bright dots on adark background. The three-dimensional posi-tions of the target can be determined from thepixel coordinates of the dots.

Original Image of Printed Target

Block CovolutionMask (BCM)

Image After Processing With Block Convolution Mask

delta x = -1.5 [mm]delta y = -0.2 [mm]delta z = -0.3 [mm]

find circuits that perform a larger num-ber of logic functions.

In order to be able to score fitness inthis way, one must ensure that circuit out-put is a digital waveform at every value ofthe control voltage, so that the outputcan be classified as a particular logic func-tion. Nevertheless, it has been observedthat the circuits generated during evolu-

tionary search typically generate analogoutputs, taking values between zero voltsand the power-supply voltage. In order tosolve this problem, the output of an evolv-ing circuit is digitized by use of a buffer,as illustrated in the figure. Whereas thedirect output of the evolving circuit isevaluated in the unmodified method, thebuffered output is evaluated in the modi-

fied method. In effect, for the purpose ofevaluation, the buffer becomes part ofany such evolved circuit.

This work was done by Adrian Stoica andRicardo Zebulum of Caltech for NASA’s JetPropulsion Laboratory. Further informa-tion is contained in a TSP (see page 1).NPO-40934


the Space Shuttle Orbiters (see Figure1). The system may be easily adapted toother applications that involve similarcoordinate-measuring requirements.Examples of such applications includemanufacturing, construction, prelimi-nary approximate land surveying, andaerial surveying.

For some applications with rectangu-lar symmetry, it is feasible and desirableto attach a target composed of blackand white squares to an object of inter-est (see Figure 2). For other situations,

where circular symmetry is more desir-able, circular targets also can be cre-ated. Such a target can readily be gen-erated and modified by use ofcommercially available software andprinted by use of a standard officeprinter. All three relative coordinates(x, y, and z) of each target can be deter-mined by processing the video image ofthe target. Because of the unique de-sign of corresponding image-processingfilters and targets, the vision-based posi-tion-measurement system is extremely

robust and tolerant of widely varyingfields of view, lighting conditions, andvarying background imagery.

This work was done by John Lane, Christo-pher Immer, Jeffrey Brink, and Robert Youngquistof Dynacs, Inc. for Kennedy Space Center.For further information, contact:

Christopher ImmerASRC AerospaceKennedy Space Center, FL 32899Phone No. (321) 867-6752

KSC-12397/67/68/442

N-Type δ Doping of High-Purity Silicon Imaging ArraysSuccess depends on details of a low-temperature MBE process.NASA’s Jet Propulsion Laboratory, Pasadena, California

A process for n-type (electron-donor)delta (δ) doping has shown promise as ameans of modifying back-illuminatedimage detectors made from n-dopedhigh-purity silicon to enable them to de-tect high-energy photons (ultraviolet andx-rays) and low-energy charged particles(electrons and ions). This process is appli-cable to imaging detectors of severaltypes, including charge-coupled devices,hybrid devices, and complementary metaloxide/semiconductor detector arrays.

Delta doping is so named because itsdensity-vs.-depth characteristic is reminis-cent of the Dirac δ function (impulsefunction): the dopant is highly concen-trated in a very thin layer. Preferably, thedopant is concentrated in one or at mosttwo atomic layers in a crystal plane and,therefore, δ doping is also known asatomic-plane doping. The use of δ dopingto enable detection of high-energy pho-tons and low-energy particles was re-ported in several prior NASA Tech Briefs ar-ticles. As described in more detail in thosearticles, the main benefit afforded by δdoping of a back-illuminated silicon de-tector is to eliminate a “dead” layer at theback surface of the silicon wherein high-energy photons and low-energy particlesare absorbed without detection. An addi-tional benefit is that the delta-doped layercan serve as a back-side electrical contact.

Delta doping of p-type silicon detec-tors is well established. The develop-ment of the present process addressesconcerns specific to the δ doping ofhigh-purity silicon detectors, which aretypically n-type. The present process in-volves relatively low temperatures, isfully compatible with other processesused to fabricate the detectors, and doesnot entail interruption of thoseprocesses. Indeed, this process can bethe last stage in the fabrication of an im-aging detector that has, in all other re-spects, already been fully processed, in-cluding metallized.

This process includes molecular-beamepitaxy (MBE) for deposition of threelayers, including metallization. The suc-cess of the process depends on accuratetemperature control, surface treatment,growth of high-quality crystalline silicon,and precise control of thicknesses of lay-ers. MBE affords the necessary nanome-ter-scale control of the placement ofatoms for delta doping.

More specifically, the process con-sists of MBE deposition of a thin siliconbuffer layer, the n-type δ doping layer,and a thin silicon cap layer. The ndopant selected for initial experimentswas antimony, but other n dopants as(phosphorus or arsenic) could be used.All n-type dopants in silicon tend to

surface-segregate during growth, lead-ing to a broadened dopant-concentra-tion-versus-depth profile. In order tokeep the profile as narrow as possible,the substrate temperature is held below300 °C during deposition of the siliconcap layer onto the antimony delta layer.The deposition of silicon includes a sil-icon-surface-preparation step, involv-ing H-termination, that enables thegrowth of high-quality crystalline sili-con at the relatively low temperaturewith close to full electrical activation ofdonors in the surface layer.

This work was done by Jordana Blacksberg,Michael Hoenk, and Shouleh Nikzad of Cal-tech for NASA’s Jet Propulsion Labora-tory. Further information is contained in aTSP (see page 1).





Software

Avionics System Architecture Tool

Avionics System Architecture Tool(ASAT) is a computer program intendedfor use during the avionics-system-archi-tecture-design phase of the process of de-signing a spacecraft for a specific mis-sion. ASAT enables simulation of thedynamics of the command-and-data-han-dling functions of the spacecraft avionicsin the scenarios in which the spacecraft isexpected to operate. ASAT is built uponI-Logix Statemate MAGNUM, providinga complement of dynamic system model-ing tools, including a graphical user in-terface (GUI), modeling checking capa-bilities, and a simulation engine. ASATaugments this with a library of prede-fined avionics components and addi-tional software to support building andanalyzing avionics hardware architec-tures using these components.

This program was written by Savio Chau,Ronald Hall, and Marcus Traylor of Caltech,and Adrian Whitfield of I-Logix 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-30629.

Updated Chemical Kineticsand Sensitivity AnalysisCode

An updated version of the GeneralChemical Kinetics and Sensitivity Analysis(LSENS) computer code has becomeavailable. A prior version of LSENS was de-scribed in “Program Helps to DetermineChemical-Reaction Mechanisms” (LEW-15758), NASA Tech Briefs, Vol. 19, No. 5(May 1995), page 66. To recapitulate:LSENS solves complex, homogeneous,gas-phase, chemical-kinetics problems(e.g., combustion of fuels) that are repre-sented by sets of many coupled, nonlinear,first-order ordinary differential equations.LSENS has been designed for flexibility,convenience, and computational effi-ciency. The present version of LSENS in-corporates mathematical models for (1) astatic system; (2) steady, one-dimensionalinviscid flow; (3) reaction behind an inci-dent shock wave, including boundary-layer correction; (4) a perfectly stirred re-actor; and (5) a perfectly stirred reactorfollowed by a plug-flow reactor. In addi-

tion, LSENS can compute equilibriumproperties for the following assignedstates: enthalpy and pressure, temperatureand pressure, internal energy and volume,and temperature and volume. For staticand one-dimensional-flow problems, in-cluding those behind an incident shockwave and following a perfectly stirred reac-tor calculation, LSENS can compute sensi-tivity coefficients of dependent variablesand their derivatives, with respect to theinitial values of dependent variablesand/or the rate-coefficient parameters ofthe chemical reactions.

This program was written by KrishnanRadhakrishnan of the Institute for Computa-tional Mechanics in Propulsion for GlennResearch Center. Further information iscontained in a TSP (see page 1).

Inquiries concerning rights for the commer-cial use of this invention should be addressedto NASA Glenn Research Center, CommercialTechnology Office, Attn: Steve Fedor, MailStop 4–8, 21000 Brookpark Road, Cleve-land, Ohio 44135. Refer to LEW-17519-1.

Predicting Flutter andForced Response in Turbomachinery

TURBO-AE is a computer code thatenables detailed, high-fidelity modelingof aeroelastic and unsteady aerodynamiccharacteristics for prediction of flutter,forced response, and blade-row interac-tion effects in turbomachinery. Flowregimes that can be modeled includesubsonic, transonic, and supersonic,with attached and/or separated flowfields. The three-dimensional Reynolds-averaged Navier-Stokes equations aresolved numerically to obtain extremelyaccurate descriptions of unsteady flowfields in multistage turbomachinery con-figurations. Blade vibration is simulatedby use of a dynamic-grid-deformationtechnique to calculate the energy ex-change for determining the aerody-namic damping of vibrations of blades.The aerodynamic damping can be usedto assess the stability of a blade row.TURBO-AE also calculates the unsteadyblade loading attributable to such exter-nal sources of excitation as incominggusts and blade-row interactions. Theseblade loadings, along with aerodynamicdamping, are used to calculate theforced responses of blades to predicttheir fatigue lives. Phase-lagged bound-ary conditions based on the direct-store

method are used to calculate nonzerointerblade phase-angle oscillations; thispractice eliminates the need to modelmultiple blade passages, and, hence, en-ables large savings in computational re-sources.

This program was written by Dale E. Van-Zante and John J. Adamczyk of Glenn Re-search Center; Rakesh Srivastava, Milind A.Bakhle, and Aamir Shabbir of the University ofToledo; Jen-Ping Chen and J. Mark Janus ofMississippi State University; Wai-Ming To ofAP Solutions, Inc.; and John Barter of GEAircraft Engines. Further information is con-tained 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-17514-1.

Upgrades of Two ComputerCodes for Analysis of Turbomachinery

Major upgrades have been made in twoof the programs reported in “Five Com-puter Codes for Analysis of Turbomachin-ery” (LEW-16851), NASA Tech Briefs, Vol.23, No. 11 (November 1999), page 28. Theaffected programs are:• Swift — a code for three-dimensional

(3D) multiblock analysis; and • TCGRID, which generates a 3D grid

used with Swift.Originally utilizing only a central-dif-

ferencing scheme for numerical solu-tion, Swift was augmented by addition oftwo upwind schemes that give greater ac-curacy but take more computing time.Other improvements in Swift include ad-dition of a shear-stress-transport turbu-lence model for better prediction of ad-verse pressure gradients, addition of anH-grid capability for flexibility in model-ing flows in pumps and ducts, and mod-ification to enable simultaneous model-ing of hub and tip clearances.Improvements in TCGRID include mod-ifications to enable generation of gridsfor more complicated flow paths and ad-dition of an option to generate gridscompatible with the ADPAC code usedat NASA and in industry. For both codes,new test cases were developed and docu-mentation was updated. Both codeswere converted to Fortran 90, with dy-namic memory allocation. Both codeswere also modified for ease of use in


both UNIX and Windows operating sys-tems.

These programs were written by Rodrick V.Chima and Meng-Sing Liou of Glenn Re-search Center. Further information is con-tained in a TSP (see page 1).

Inquiries concerning rights for the commer-cial use of this invention should be addressedto NASA Glenn Research Center, CommercialTechnology Office, Attn: Steve Fedor, MailStop 4–8, 21000 Brookpark Road, Cleveland,Ohio 44135. Refer to LEW-17635/88-1.

Program Facilitates CMMIAppraisals

A computer program has been writtento facilitate appraisals according to themethodology of Capability MaturityModel Integration (CMMI). [CMMI is agovernment/industry standard, main-tained by the Software Engineering Insti-tute at Carnegie Mellon University, forobjectively assessing the engineering ca-pability and maturity of an organization(especially, an organization that producessoftware)]. The program assists in prepa-ration for a CMMI appraisal by providingdrop-down lists suggesting required arti-facts or evidence. It identifies processareas for which similar evidence is re-quired and includes a copy feature thatreduces or eliminates repetitive dataentry. It generates reports to show the en-tire framework for reference, the ap-praisal artifacts to determine readinessfor an appraisal, and lists of intervieweesand questions to ask them during the ap-praisal. During an appraisal, the programprovides screens for entering observa-tions and ratings, and reviewing evidenceprovided thus far. Findings concerningstrengths and weaknesses can be ex-ported for use in a report or a graphicalpresentation. The program generates achart showing capability level ratings ofthe organization. A context-sensitive Win-dows help system enables a novice to usethe program and learn about the CMMIappraisal process.

This program was written by Wesley Sweetser ofGoddard Space Flight Center. Further infor-mation is contained in a TSP (see page 1).GSC-14782-1

Grid Visualization ToolThe Grid Visualization Tool (GVT) is

a computer program for displaying thepath of a mobile robotic explorer(rover) on a terrain map. The GVTreads a map-data file in either portablegraymap (PGM) or portable pixmap(PPM) format, representing a gray-scale

or color map image, respectively. TheGVT also accepts input from path-plan-ning and activity-planning software.From these inputs, the GVT generates amap overlaid with one or more roverpath(s), waypoints, locations of targetsto be explored, and/or target-status in-formation (indicating success or failurein exploring each target). The displaycan also indicate different types of pathsor path segments, such as the path actu-ally traveled versus a planned path orthe path traveled to the present positionversus planned future movement alonga path. The program provides for updat-ing of the display in real time to facili-tate visualization of progress. The size ofthe display and the map scale can bechanged as desired by the user. TheGVT was written in the C++ languageusing the Open Graphics Library(OpenGL) software. It has been com-piled for both Sun Solaris and Linux op-erating systems.

This program was written by CarolineChouinard, Forest Fisher, Tara Estlin,Daniel Gaines, and Steven Schaffer of Cal-tech for NASA’s Jet Propulsion Labora-tory. Further information is contained in aTSP (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-40303.

Program Computes SoundPressures at RocketLaunches

Launch Vehicle External SoundPressure is a computer program thatpredicts the ignition overpressure andthe acoustic pressure on the surfacesand in the vicinity of a rocket andlaunch pad during launch. The pro-gram generates a graphical user inter-face (GUI) that gathers input datafrom the user. These data include thecritical dimensions of the rocket and ofany launch-pad structures that may actas acoustic reflectors, the size andshape of the exhaust duct or flame de-flector, and geometrical and opera-tional parameters of the rocket engine.For the ignition-overpressure calcula-tions, histories of the chamber pres-sure and mass flow rate also are re-quired. Once the GUI has gathered theinput data, it feeds them to ignition-overpressure and launch-acoustics rou-tines, which are based on several ap-proximate mathematical models ofdistributed sources, transmission, andreflection of acoustic waves. The out-

put of the program includes ignitionoverpressures and acoustic pressures atspecified locations.

This program was written by Gary Ogg,Roy Heyman, Michael White, and KarlEdquist of Applied Research Associates, Inc.,for Marshall Space Flight Center. For fur-ther information, contact the company atwww.ara.com.MFS-31568

Solar-System EphemerisToolbox

NASA’s Jet Propulsion Laboratory (JPL)generates planetary and lunar ephemerisdata and FORTRAN routines that allowusers to obtain state data for the Sun, themoon, and the planets. The JPL Solar Sys-tem Ephemeris Toolbox, developed atKennedy Space Center, is a set of functionsthat provides the same functionality in theMATLAB computing environment alongwith some additional capabilities. The tool-box can be used interactively via a graphicaluser interface (GUI), or individual func-tions can be called from the MATLAB com-mand prompt or other MATLAB scriptsand functions. The toolbox also includesutility functions to define and perform co-ordinate transformation (e.g., mean-of-date, true-of-date, J2000) that are commonin the use of these ephemerides. An at-tached README file guides the userthrough the process of constructing binaryephemeris files, verifying correct installa-tion, and using functions to extract statedata. This process also can be performedusing the GUI. Help from each toolboxfunction is available through MATLAB’s“help” function. Many of the functions inthe toolbox are MATLAB equivalents of theJPL-written FORTRAN programs and sub-routines used for the same purposes. Anovice can use the GUI to extract state data,while a more experienced user can use thefunctions directly, as needed, in his/her ap-plications. The toolbox has been testedusing MATLAB Releases 13 and 14.

This program was written by Charles F. Walkerof Kennedy Space Center. For further infor-mation, access www.openchannelsoftware.org.KSC-12544

Data-Acquisition Softwarefor PSP/TSP Wind-TunnelCameras

Wing-Viewer is a computer program foracquisition and reduction of image data ac-quired by any of five different scientific-grade commercial electronic cameras usedat Langley Research center to observewind-tunnel models coated with pressure-


or temperature-sensitive paints (PSP/TSP).Wing-Viewer provides full automation ofcamera operation and acquisition ofimage data, and has limited data-prepro-cessing capability for quick viewing of theresults of PSP/TSP test images. Wing-Viewer satisfies a requirement for a stan-dard interface between all the cameras

and a single personal computer: Writtenby use of Microsoft Visual C++ and the Mi-crosoft Foundation Class Library as aframework, Wing-Viewer has the ability tocommunicate with the C/C++ software li-braries that run on the controller circuitcards of all five cameras. Wing-Viewersaves image data in tagged image file

(TIF) version 6.0 format. Wing-Viewercan function on computers that run anyof the several Windows operating systems,including Windows 95, 98, 2000, and NT.

This program was written by Tahani R.Amer and William K. Goad of Langley Re-search Center. Further information is con-tained in a TSP (see page 1). LAR-16474-1


Comparative tests have been per-formed to evaluate the corrosion-preven-tion capabilities of an experimental paintof the type described in “Water-Borne,Silicone-Based, Primerless Paints,” NASATech Briefs, Vol. 26, No. 11 (November2002), page 30. To recapitulate: thesepaints contain relatively small amountsof volatile organic solvents and were de-veloped as substitutes for traditional anti-corrosion paints that contain largeamounts of such solvents. An additionaldesirable feature of these paints is thatthey can be applied without need forprior application of primers to ensureadhesion.

The test specimens included panels ofcold-rolled steel, stainless steel 316, and alu-minum 2024-T3. Some panels of each ofthese alloys were left bare and some werecoated with the experimental water-borne,silicone-based, primerless paint. In addition,some panels of aluminum 2024-T3 andsome panels of a fourth alloy (stainless steel304) were coated with a commercial solvent-borne paint containing aluminum and zincflakes in a nitrile rubber matrix. In the tests,the specimens were immersed in an aerated3.5-weight-percent aqueous solution ofNaCl for 168 hours. At intervals of 24 hours,the specimens were characterized by elec-trochemical impedance spectroscopy (EIS)and measurements of corrosion potentials.The specimens were also observed visually.

As indicated by photographs of speci-mens taken after the 168-hour immersion

(see figure), the experimental primerlesssilicone paint was effective in preventingcorrosion of stainless steel 316, but failed toprotect aluminum 2024-T3 and cold-rolledsteel. The degree of failure was greater inthe case of the cold-rolled steel. On thebasis of visual observations, EIS, and corro-sion-potential measurements, it was con-cluded that the commercial aluminum-and zinc-filled nitrile rubber coating affords

superior corrosion protection to aluminum2024-T3 and is somewhat less effective inprotecting stainless steel 304.

This work was done by Luz Marina Calleand Louis G. MacDowell of Kennedy SpaceCenter, and Rubie D. Vinje of ASRC Aero-space. For further information, contact theKennedy Innovative Partnerships Office at(321) 867-1463.KSC-12520

Materials

(a) (b)

(c) (d)

1 cm

1 cm

1 cm

Corrosion-Prevention Capabilities of a Water-Borne, Silicone-Based, Primerless CoatingSome formulations are better for steel, some for aluminum.John F. Kennedy Space Center, Florida

Blistering of an Experimental Silicone Paint is manifest on two alloy specimens after immersion for aweek in an aerated saltwater solution: (a) silicone-coated aluminum 2024-T3 panel, (b) silicone-coated316 stainless-steel panel, (c) silicone-coated cold-rolled-steel panel, and (d) aluminum 2024-T3 panelcoated with aluminum- and zinc-filled nitrile rubber.

A sol-gel process has been developedas a superior alternative to a priorprocess for making platinum-rutheniumalloy catalysts for electro-oxidation ofmethanol in fuel cells. The starting mate-rials in the prior process are chloridesalts of platinum and ruthenium. The

process involves multiple steps, is time-consuming, and yields a Pt-Ru productthat has relatively low specific surfacearea and contains some chloride residue.Low specific surface area translates to in-complete utilization of the catalytic activ-ity that might otherwise be available,

while chloride residue further reducescatalytic activity (“poisons” the catalyst).In contrast, the sol-gel process involvesfewer steps and less time, does not leavechloride residue, and yields a product ofgreater specific area and, hence, greatercatalytic activity.

Sol-Gel Process for Making Pt-Ru Fuel-Cell CatalystsRelative to another process, this one takes less time and yields better results.NASA’s Jet Propulsion Laboratory, Pasadena, California


In this sol-gel process (see figure), thestarting materials are platinum(II) acety-lacetonate [Pt(C5H7O2)2, also denotedPt-acac] and ruthenium(III) acetylaceto-nate [Ru(C5H7O2)3, also denoted Ru-acac]. First, Pt-acac and Ru-acac are dis-

solved in acetone at the desired concen-trations (typically, 0.00338 moles of eachsalt per 100 mL of acetone) at a temper-ature of 50 °C. A solution of 25 percenttetramethylammonium hydroxide[(CH3)4NOH, also denoted TMAH] inmethanol is added to the Pt-acac/Ru-acac/acetone solution to act as a high-molecular-weight hydrolyzing agent.The addition of the TMAH counteractsthe undesired tendency of Pt-acac andRu-acac to precipitate as separate phasesduring the subsequent evaporation ofthe solvent, thereby helping to yield adesired homogeneous amorphous gel.The solution is stirred for 10 minutes,then the solvent is evaporated until thesolution becomes viscous, eventuallytransforming into a gel. The viscous gelis dried in air at a temperature of 170 °Cfor about 10 hours. The dried gel iscrushed to make a powder that is the im-mediate precursor of the final catalyticproduct.

The precursor powder is converted tothe final product in a controlled-atmos-phere heat treatment. Desirably, thefinal product is a phase-pure (Pt phaseonly) Pt-Ru powder with a high specificsurface area. The conditions of the con-trolled-atmosphere heat are critical forobtaining the aforementioned desired

properties. A typical heat treatment thatyields best results for a catalytic alloy ofequimolar amounts of Pt and Ru con-sists of at least two cycles of heating to atemperature of 300 °C and holding at300 °C for several hours, all carried outin an atmosphere of 1 percent O2 and99 percent N2. The resulting powderconsists of crystallites with typical lineardimensions of <10 nm. Tests have shownthat the powder is highly effective in cat-alyzing the electro-oxidation ofmethanol.

This work was done by SekharipuramNarayanan and Thomas Valdez of Caltech,and Prashant Kumta and Y. Kim ofCarnegie-Mellon University for NASA’s JetPropulsion Laboratory. Further informa-tion is contained in a TSP (see page 1).


Innovative Technology Assets Management:JPLMail Stop 202-2334800 Oak Grove DrivePasadena, CA 91109-8099(818) 354-2240E-mail: [email protected]

Refer to NPO-30500, volume and number of thisNASA Tech Briefs issue, and the page number.

Pt-acac Ru-acac

Dissolve at 50 °C in Acetone

Crush Dried Gel to Powder

Heat in Controlled Atmosphere

Test Final Product

Evaporate the Solution

Add (CH3)4NOH

Pt-Ru Catalytic Powder Is Made from organicsalts of Pt and Ru in a sol-gel process that in-volves fewer steps and less time than does aprocess based on chloride salts of Pt and Ru.

Making Activated Carbon for Storing GasLyndon B. Johnson Space Center, Houston, Texas

Solid disks of microporous activated car-bon, produced by a method that enablesoptimization of pore structure, have beeninvestigated as means of storing gas (espe-cially hydrogen for use as a fuel) at rela-tively low pressure through adsorption onpore surfaces. For hydrogen and othergases of practical interest, a narrow distribu-tion of pore sizes <2 nm is preferable. Thepresent method is a variant of a previouslypatented method of cyclic chemisorptionand desorption in which a piece of carbon

is alternately (1) heated to the lower of twoelevated temperatures in air or other oxi-dizing gas, causing the formation of stablecarbon/oxygen surface complexes; then(2) heated to the higher of the two elevatedtemperatures in flowing helium or otherinert gas, causing the desorption of the sur-face complexes in the form of carbonmonoxide. In the present method, porestructure is optimized partly by heating to atemperature of 1,100 °C during carboniza-tion. Another aspect of the method exploits

the finding that for each gas-storage pres-sure, gas-storage capacity can be maxi-mized by burning off a specific proportion(typically between 10 and 20 weight per-cent) of the carbon during the cyclicchemisorption/desorption process.

This work was done by Marek A. Wójtowiczand Michael A. Serio of Advanced Fuels Re-search, Inc., and Eric M. Suuberg (consultant)for Johnson Space Center. For further informa-tion, contact the Johnson Innovative PartnershipsOffice at (281) 483-3809. MSC-23233


Machinery/Automation

Five-Axis, Three-Magnetic-Bearing Dynamic Spin Rig Higher-order vibrational modes can be excited and higher rotational speeds attained. John H. Glenn Research Center, Cleveland, Ohio

The Five-Axis, Three-Magnetic-BearingDynamic Spin Rig is an apparatus for vi-bration testing of turbomachine blades ina vacuum at rotational speeds from 0 to40,000 rpm. This rig (see figure) includes(1) a vertically oriented shaft on which ismounted an assembly comprising a rotorholding the blades to be tested, (2) twoactively controlled heteropolar radialmagnetic bearings at opposite ends of theshaft, and (3) an actively controlled mag-netic thrust bearing at the upper end ofthe shaft. This rig is a more capable suc-cessor to a prior apparatus, denoted theDynamic Spin Rig (DSR), that included avertically oriented shaft with a mechani-cal thrust bearing at the upper end and asingle actively controlled heteropolar ra-dial magnetic bearing at the lower end.

The five-axis, three-magnetic-bearingconfiguration of the present rig enablesfull magnetic suspension of the rotor —eliminating mechanical contact betweenthe rotor and the bearings during opera-tion. Whereas frictional heating in themechanical thrust bearing of the priorDSR made it necessary to limit rotationalspeed to 18,000 rpm or less, the absenceof frictional heating in the present rigmakes it possible to operate at higherspeed, provided that a rotor of appropri-ate high-speed design is installed.

In the prior DSR, it was not possibleto excite vibrations in higher-ordermodes in bladed-disk test assemblies by

Rig SupportStructure

Speed-PickupWheel

DisplacementSensor

Displacement Sensor(Shown Out ofTrue Position)

UpperThrust-Bearing

Coil

Stator RadialBearing Coil Upper

Magnetic RadialBearingStator Radial Bearing

Laminations

Rig SupportStructure

Rotor Radial BearingLaminations

LowerThrust-Bearing

Coil

Disk

Shaft

Blade

Thrust PlateMagneticThrustBearing

LowerRadial-Bearing

HousingLowerMagnetic RadialBearing

MechanicalRadial Touchdown Bearing

Rotational-SpeedSensors

MechanicalTouchdown Bearing(Radial and Thrust)

Upper Radial-BearingHousing

The Five-Axis, Three-Magnetic-Bearing configuration of this dynamic spin rig makes it possible to excitehigh-order vibrational modes of the disk-and-blade assembly. The five axes are the vertical thrust axis of thethrust bearing and two mutually perpendicular horizontal force axes for each of the two radial bearings.

System Regulates the Water Contents of Fuel-Cell StreamsLyndon B. Johnson Space Center, Houston, Texas

An assembly of devices provides forboth humidification of the reactant gasstreams of a fuel cell and removal of theproduct water (the water generated byoperation of the fuel cell). The assemblyincludes externally-sensing forward-pres-sure regulators that supply reactantgases (fuel and oxygen) at variable pres-sures to ejector reactant pumps. Theejector supply pressures depend on theconsumption flows. The ejectors developdifferential pressures approximately

proportional to the consumption flowrates at constant system pressure andwith constant flow restriction betweenthe mixer-outlet and suction ports of theejectors. For removal of product waterfrom the circulating oxygen stream, theassembly includes a water/gas separatorthat contains hydrophobic and hydrophilicmembranes. The water separator im-poses an approximately constant flow re-striction, regardless of the quality of thetwo-phase flow that enters it from the

fuel cell. The gas leaving the water sepa-rator is nearly 100 percent humid. Thisgas is returned to the inlet of the fuel cellalong with a quantity of dry incomingoxygen, via the oxygen ejector, therebyproviding some humidification.

This work was done by Arturo Vasquezand Scott Lazaroff of Johnson Space Cen-ter. For further information, contact theJohnson Innovative Partnerships Office at(281) 483-3809.MSC-23079


applying feed-forward control excita-tions to the magnetic bearings. This lim-itation was due partly to the lateral con-straint imposed on the rotor by themechanical bearing at the upper end ofthe shaft, partly to the direct effect offriction in that bearing, and partly tothe aforementioned speed limit im-posed to prevent excessive frictionalheating. In contrast, by virtue of thefully magnetic nature of the present sus-pension, there is no lateral constraintagainst vibrations, and excitation ampli-

tudes can be greater than in the priorDSR. By applying appropriate feed-for-ward bounce-mode and tilt-mode con-trol excitation command to the activemagnetic bearings in the present rig,one can excite vibrations in a variety ofmodes. The combination of large-ampli-tude feed-forward excitation and higherrotational speed makes it possible to ex-cite higher-order vibrations in a bladed-disk test assembly.

This work was done by Carlos R. Morrison,Andrew Provenza, Anatole Kurkov, Oral

Mehmed, and Dexter Johnson of Glenn Re-search Center; Gerald Montague of TheArmy Research Laboratory; and KirstenDuffy and Ralph Jansen of The Universityof Toledo. Further information is containedin a TSP (see page 1).

Inquiries concerning rights for the commer-cial use of this invention should be addressedto NASA Glenn Research Center, InnovativePartnerships Office, Attn: Steve Fedor, MailStop 4–8, 21000 Brookpark Road, Cleveland,Ohio 44135. Refer to LEW-17757-1.


Manufacturing

Modifications of Fabrication of Vibratory MicrogyroscopesThe goal is to increase production yields.NASA’s Jet Propulsion Laboratory, Pasadena, California

A micromachining process for the fab-rication of vibratory microgyroscopesfrom silicon wafers, and aspects of themicrogyroscope design that are inextri-cably linked with the fabricationprocess, have been modified in an effortto increase production yields from per-spectives of both quantity and quality.

Prior to the modifications, the effectiveproduction yield of working microgyro-scopes was limited to one or less perwafer. The modifications are part of acontinuing effort to improve the designand increase production yields to morethan 30 working microgyroscopes perwafer.

A discussion of pertinent aspects of theunmodified design and the unmodifiedfabrication process is prerequisite to ameaningful description of the modifica-tions. The design of the microgyroscopepackage was not conducive to high yieldand rapid testing of many microgyro-scopes. One of the major impediments tohigh yield and testing was found to lie in vi-bration-isolation beams around the fouredges of each microgyroscope, whichbeams were found to be unnecessary forachieving high resonance quality factors(Q values) characterizing the vibrations ofpetallike cantilevers.

The fabrication process included an 8-µm-deep plasma etch. The purpose of theetch was to create 8-µm vertical gaps,below which were to be placed large goldevaporated electrodes and sensing pads todrive and sense resonant vibrations of the“petals.” The process also included a stepin which bridges between dies were cut toseparate the dies.

The etched areas must be kept cleanand smooth (free of debris and spikes),because any object close to 8 µm high inthose areas would stop the vibrations.However, it was found that after the etch,there remained some spikes with heightsthat were, variously, almost as high or ashigh as the etch depth. It also was foundthat the cutting of bridges created silicondebris, some of which lodged in the 8-µmgaps and some of which landed on top ofthe petals. The masses added to thepetals by the debris altered resonance fre-quencies and/or Q values to unaccept-able degrees. Hence, the spikes and thedebris have been conjectured to causemost of the observed malfunctions ofnewly fabricated microgyroscopes.

Another pertinent aspect of the un-modified design and process was the fab-rication of electrodes and the 8-µm ca-pacitance gap on a 500-µm-thick wafer,and the fabrication of a 3-mm-thickbaseplate from another wafer. It was nec-essary to bond these wafers to eachother in an assembly step that was laterfound to be superfluous in that it couldbe eliminated by a suitable modificationof the design.

The modifications include a redesign

Zero-Insertion- Force Socket

Microgyroscope

Ceramic Package

Circuit Board

MICROGYROSCOPE IN CERAMIC PACKAGE NEXT TO SOCKET

PACKAGED GYROSCOPE PLUGGED INTO SOCKET FOR TESTING

A Plug-and-Test Design enables the testing of many microgyroscopes in a short time.


of the microgyroscope package to elimi-nate the vibration-isolation beams whileproviding acceptably high Q values (≈4× 104). The modified design includes aplug-in feature for quick testing (see fig-ure). The plasma etch has been replacedby a wet etch, using a specially formu-lated KOH-based solution, that does notleave spikes. The design of the bridgeshas been modified to incorporate doublenotches, such that they can be cut with-out producing much debris, and a spe-cial suction tool resembling one used bya dentist has been developed to collectflying debris during cutting.

The superfluous assembly step hasbeen eliminated by modifying the de-

sign so that all the functional parts previ-ously fabricated on the 500-µm and 3-mm wafers are now fabricated entirelyon 3-mm baseplate wafers only. In a pre-vious approach to elimination of the su-perfluous step, KOH etches were madethrough 3-mm wafers, then metal pat-terns were formed by evaporating metalswhile using shadow masks (not standardpractice). In the modified process, themetals are evaporated first (standardpractice), then holes are ground by useof a diamond-tipped drill on an indextable.

This work was done by Sam Y. Bae, KarlY. Yee, and Dean Wiberg of Caltech forNASA’s Jet Propulsion Laboratory. Fur-

ther information is contained in a TSP (seepage 1).





Bio-Medical

Electroporation System for Sterilizing WaterAmounts of chemicals needed for sterilization are reduced.Lyndon B. Johnson Space Center, Houston, Texas

A prototype of an electroporation sys-tem for sterilizing wastewater or drink-ing water has been developed. In elec-troporation, applied electric fieldscause transient and/or permanentchanges in the porosities of living cells.Electroporation at lower field strengthscan be exploited to increase the effi-ciency of chemical disinfection (as inchlorination). Electroporation athigher field strengths is capable of inac-tivating and even killing bacteria andother pathogens, without use of chemi-cals. Hence, electroporation is at least apartial alternative to chlorination.

The transient changes that occur inmicro-organisms at lower electric-fieldstrengths include significantly increaseduptake of ions and molecules. Such in-creased uptake makes it possible toachieve disinfection at lower doses ofchemicals (e.g., chlorine or ozone) thanwould otherwise be needed. Lowerdoses translate to lower costs and re-duced concentrations of such carcino-genic chemical byproducts astrichloromethane. Higher electric fieldscause cell membranes to lose semiper-meability and thereby become unable tofunction as selective osmotic barriers be-tween the cells and the environment.This loss of function is the cause of thecell death at higher electric-field intensi-ties. Experimental evidence does not in-dicate cell lysis but, rather, combinedleaking of cell proteins out of the cells as

well as invasion of foreign chemicalcompounds into the cells.

The concept of electroporation is notnew: it has been applied in molecularbiology and genetic engineering fordecades. However, the laboratory-scaleelectroporators used heretofore havebeen built around small (400-micro-liter) cuvettes, partly because the small-ness facilitates the generation of electricfields of sufficient magnitude to causeelectroporation. Moreover, most labora-tory-scale electroporators have been de-

signed for testing static water. In con-trast, the treatment cell in the presentsystem is much larger and features aflow-through geometry, such that elec-tric fields strong enough to effect 99.9-percent disinfection can be applied towater flowing in a pipe.

The figure schematically depicts oneversion of the prototype system, whereinthe output of a high-voltage pulse genera-tor is applied to two electrodes on oppo-site sides of a flow-through electropora-tion cell. The pulse amplitude, duration,

Water Reservoir

Timer:(1 Hz,20 ms

Typical)

Start/Stop

Opto-electronic

Relay

Voltage Set

Power Supply(10 kV, 2 kW)

High-Voltage Pulse Generator

220 VAC

120 VAC

+

–

ElectromagneticRelay Switch

Pump

Flow-ThroughElectroporation Cell

Electrodes

High-Voltage Pulses Are Applied to a pair of electrodes as water flows between them. Depending on thepulse amplitude, the resulting electric field between the electrodes either makes pathogens in the watermore vulnerable to a disinfecting chemical or else inactivates them even in the absence of such a chemical.

A chamber has been designed to enablegrowth and observation of microcoloniesof fungi in isolation from the external en-vironment. Unlike prior fungus-growingapparatuses, this chamber makes it possi-ble to examine a fungus culture withoutdisrupting it. Partly resembling a small pic-ture frame, the chamber includes a metalplate having a rectangular through-the-thickness opening with recesses for a topand a bottom cover glass, an inlet for air,

and an inlet for water. The bottom coverglass is put in place and held there by clips,then a block of nutrient medium and amoisture pad are placed in the opening.The block is inoculated, then the topcover glass is put in place and held thereby clips. Once growth is evident, the cham-ber can be sealed with tape. Little (if any)water evaporates past the edges of thecover glasses, and, hence there is little (ifany) need to add water. A microscope can

be used to observe the culture through ei-ther cover glass. Because the culture issealed in the chamber, it is safe to examinethe culture without risking contamination.The chamber can be sterilized and reused.

This work was done by Duane L. Piersonof Johnson Space Center and Thomas C.Molina of KRUG Life Sciences. For furtherinformation, contact the Johnson InnovativePartnerships Office at (281) 483-3809.MSC-22904

Chamber for Growing and Observing FungiLyndon B. Johnson Space Center, Houston, Texas


and repetition period are chosen to ob-tain the desired degree of disinfection.Most critical is the amplitude, which ischosen in consideration of the interelec-trode gap (1 cm in the prototype) to ob-tain the needed electric-field intensity.The threshold electric-field intensity fortransient changes in permeability and re-duced-dosage infection is about 0.2kV/cm; the threshold for inactivation isabout 5 kV/cm. In a practical system, theelectroporation cell would be equippedwith multiple pairs of electrodes along theflow path and the high-voltage pulses ap-

plied to the pairs would be synchronizedso that any given small volume of waterwould be subjected to multiple high-volt-age pulses on its way through the electro-poration cell.

Electroporation sterilization technologyis best employed in small point-of-entry(POE) and point-of-use (POU) applica-tions as in homes or other small facilities.In smaller pipe diameters, it can be verycost effective, but the power usage becomesexcessive in larger water or wastewatertreatment facilities. Bioelectromagnetics,however, has developed an alternative

electromagnetic field technology that isvery cost effective in large water/waste-water treatment installations.

This work was done by Kenneth J. Schlager ofBioelectromagnetics, Inc. for Johnson SpaceCenter. For further information, contact

Kenneth J. Schlager, PresidentBioelectromagnetics, Inc.12825 Elmwood RoadElm Grove, WI 53122Phone: (262) 782-2048Fax: (262) 786-1491E-mail: [email protected]

Refer to MSC-23377.


A proposed thermoelectric devicewould exploit natural temperature dif-ferences between air and soil to harvestsmall amounts of electric energy. Be-cause the air/soil temperature differ-ence fluctuates between nighttime anddaytime, it is almost never zero, and sothere is almost always some energyavailable for harvesting. Unlike photo-voltaic cells, the proposed device couldoperate in the absence of sunlight. Un-like a Stirling engine, which could bedesigned to extract energy from theair/soil temperature difference, theproposed device would contain no mov-ing parts. The main attractive feature ofthe proposed device would be high reli-ability. In a typical application, this de-vice would be used for low-power charg-ing of a battery that would, in turn,supply high power at brief, infrequentintervals for operating an instrumenta-tion package containing sensors andcommunication circuits.

The device (see figure) would includea heat exchanger buried in soil and con-nected to a heat pipe extending up to a short distance above the ground sur-face. A thermoelectric microgenerator(TEMG) would be mounted on top ofthe heat pipe. The TEMG could be of anadvanced type, now under development,that could maintain high (relative toprior thermoelectric generators) powerdensities at small temperature differen-tials. A heat exchanger exposed to theair would be mounted on top of the

TEMG. It would not matter whether theair was warmer than the soil or the soilwarmer than the air: as long as there wasa nonzero temperature difference, heat

would flow through the device and elec-tricity would be generated.

A study of factors that could affect thedesign and operation of the device hasbeen performed. These factors includethe thermal conductances of the soil,the components of the device, the con-tacts between the components of the de-vice, and the interfaces between the heatexchangers and their environments.The study included experiments thatwere performed on a model of the de-vice to demonstrate feasibility. Because aTEMG suitable for this device was notavailable, a brass dummy componenthaving a known thermal conductance of1.68 W/K was substituted for the TEMGin the models to enable measurement ofheat flows. The model included a water-based heat pipe 30 in. (76.2 cm) longand 1 in. (2.54 cm) in diameter,wrapped with polyethylene insulation toreduce radial heat flow. Several differentside heat exchangers were tested. Onthe basis of the measurements, it waspredicted that if a prototype of the de-vice were equipped with a TEMG, dailytemperature fluctuations would cause itsoutput power to fluctuate between 0 andabout 0.1 mW, peaking to 0.35 mW dur-ing early afternoon.

This work was done by Jeffrey Snyder, Jean-Pierre Fleurial, and Eric Lawrence of Caltechfor NASA’s Jet Propulsion Laboratory.Further information is contained in a TSP(see page 1).NPO-30831

Physical Sciences

Thermoelectric Air/Soil Energy-Harvesting DeviceSmall amounts of power would be extracted from natural temperature differences.NASA’s Jet Propulsion Laboratory, Pasadena, California

HeatPipe Thermal

Insulation

Air-SideHeat Exchanger

TEMG

Air

SoilSoil

Ground-SideHeat Exchanger

Ground-SideHeat Exchanger

This Simple, Reliable Thermoelectric Devicewould harvest electric energy from the differ-ence in temperature between air and soil.

Flexible metal-fabric radiators havebeen considered as alternative meansof dissipating excess heat from space-craft and space suits. The radiators alsomay be useful in such special terrestrialapplications as rejecting heat fromspace-suit-like protective suits worn inhot work environments. In addition toflexibility and consequent ease of de-ployment and installation on objects of

varying sizes and shapes, the main ad-vantages of these radiators over con-ventional rigid radiators are that theyweigh less and occupy less volume for agiven amount of cooling capacity. A ra-diator of this type includes conven-tional stainless-steel tubes carrying acoolant fluid. The main radiating com-ponent consists of a fabric of inter-woven aluminum-foil strips bonded to

the tubes by use of a proprietaryprocess. The strip/tube bonds arestrong and highly thermally conduc-tive. Coolant is fed to and from thetubes via flexible stainless-steel mani-folds designed to accommodate flexingof, and minimize bending forces on,the fabric. The manifolds are sized tominimize pressure drops and distributethe flow of coolant evenly to all the

Flexible Metal-Fabric RadiatorsLyndon B. Johnson Space Center, Houston, Texas


tubes. The tubes and manifolds areconfigured in two independent flowloops for operational flexibility andprotective redundancy.

This work was done by Cynthia Cross andHai D. Nguyen of Johnson Space Center,and Warren Ruemmele, Kambiz K. Andish,and Sean McCalley of Lockheed Martin

Corp. For further information, contact theJohnson Innovative Partnerships Office at(281) 483-3809.MSC-23331

SecondaryMirror

Phase RetrievalCamera

Shack-HartmanCamera

LinearMechanism

LinearMechanism

CoverMechanism

DetectorDetector Lenslet

CollimatingLens

CameraAssembly

FilterWheel

OpticalFiber

LEDIlluminator

Shutter

NanolaminatePrimary Mirror

SiC SupportingStructure

Tertiary Mirror

CoverMirror

Telescope

The Design of the AHMT will utilize advanced materials and advanced sensing and control techniques to obtain imaging. The primary mirror will have a diameterof 0.75 m and an areal density less than 10 kg/m2.

The figure depicts the planned Actu-ated Hybrid Mirror Telescope (AHMT),which is intended to demonstrate a newapproach to the design and constructionof wide-aperture spaceborne telescopesfor astronomy and Earth science. Thistechnology is also appropriate for Earth-based telescopes.

The new approach can be broadlysummarized as using advanced light-weight mirrors that can be manufacturedrapidly at relatively low cost. More specif-ically, it is planned to use precise repli-cated metallic nanolaminate mirrors toobtain the required high-quality opticalfinishes. Lightweight, dimensionally sta-ble silicon carbide (SiC) structures willsupport the nanolaminate mirrors in therequired surface figures. To enable dif-

fraction-limited telescope performance,errors in surface figures will be correctedby use of mirror-shape-control actuatorsthat will be energized, as needed, by awave-front-sensing and control system.

The concepts of nanolaminate materi-als and mirrors made from nanolami-nate materials were discussed in severalprevious NASA Tech Briefs articles.Nanolaminates constitute a relativelynew class of materials that can approachtheoretical limits of stiffness andstrength. Nanolaminate mirrors are syn-thesized by magnetron sputter deposi-tion of metallic alloys and/or com-pounds on optically precise mastersurfaces to obtain optical-quality reflec-tor surfaces backed by thin shell struc-tures. As an integral part of the deposi-

tion process, a layer of gold that will con-stitute the reflective surface layer is de-posited first, eliminating the need for asubsequent and separate reflective-coat-ing process. The crystallographic tex-tures of the nanolaminate will be con-trolled to optimize the performance ofthe mirror. The entire depositionprocess for making a nanolaminate mir-ror takes less than 100 hours, regardlessof the mirror diameter.

Each nanolaminate mirror will bebonded to its lightweight SiC support-ing structure. The lightweight nanolam-inate mirrors and SiC supporting struc-tures will be fabricated from reusablemaster molds. The mirror-shape-controlactuators will be low-power, high-capaci-tance lead magnesium niobate elec-

Actuated Hybrid Mirror TelescopeA new type of lightweight, wide-aperture, precise telescope is under development.NASA’s Jet Propulsion Laboratory, Pasadena, California


trostrictive actuators that will be embed-ded in the SiC structures. The mode ofoperation of these actuators will be suchthat once power was applied, they willchange in length and once power wasremoved, they will maintain dimen-sional stability to nanometer precision.This mode of operation will enable theuse of low-power, minimally complexelectronic control circuitry.

The wave-front-sensing and controlsystem will be designed and built accord-

ing to a two-stage architecture. The firststage will be implemented by a Shack-Hartmann (SH) sensor subsystem,which will provide a large capture range.The second, higher-performance stagewill be implemented by an image-basedwave-front-sensing subsystem that will in-clude a phase-retrieval camera (PRC),and will utilize phase retrieval and othertechniques to measure wavefront errordirectly. Phase retrieval is a process inwhich multiple images of an unresolved

object are iterated to estimate the phaseof the optical system that acquired theimages. The combination of SH andphase-retrieval sensors will afford thevirtues of both a dynamic range of 105

and an accuracy of <10 nm.This work was done by Gregory Hickey,

David Redding, Andrew Lowman, DavidCohen, and Catherine Ohara of Caltech forNASA’s Jet Propulsion Laboratory. Furtherinformation is contained in a TSP (see page 1).NPO-40105

Optical Design of an Optical Communications TerminalThis airborne system would keep itself aimed at a ground station.NASA’s Jet Propulsion Laboratory, Pasadena, California

An optical communications terminal(OCT) is being developed to enable trans-mission of data at a rate as high as 2.5Gb/s, from an aircraft or spacecraft to aground station. In addition to transmittinghigh data rates, OCT will also be capableof bidirectional communications. TheOCT is meant to incorporate all of the de-sign features of a prior apparatus denotedthe Optical Communications Demonstra-tor (OCD), plus some improvements.

Like the OCD, the OCT would utilizea single telescope aperture for bothtransmitting and receiving. Also as in theOCD, a fine-steering mirror (FSM)would be included in the transmittingoptical train.

The OCT design utilizes a 1,550-nmfiber-optic amplifier transmitter like that

used in the telecommunications indus-try. Such an amplifier includes a single-mode oscillator, to which one can applymodulation such that the laser light em-anating from the fiber can convey dataat a rate in the gigabit-per-second range.The laser beam from each such ampli-fier would be coupled, via a collimatinginterface module, to a transceiver opti-cal assembly, major optical componentsof which are shown in the figure.

The OCT shall include large-field-of-view focal planes for receiving opticalcommunications and for sensing remotebeacon lasers for controlled acquisition,tracking, and pointing (in other words,beacons toward which the OCT wouldbe aimed for transmitting or receiving).The OCT could be connected to a gim-

bal assembly that could be used forcoarse aiming.

The OCT would utilize six opticalchannels — three for transmitting, threefor receiving. The transmitting channelswould be the following:• A channel for a 1,550-nm-wavelength

laser beam, which would be the maindata-modulated beam to be transmit-ted via the telescope;

• A channel for part of a split 980-nmlaser beam used as a reference beamfor fine-pointing servo control; and

• A channel for the other part of thesplit 980-nm beam used for calibrationof a coarse-acquisition charge-coupleddevice (CCD).The receiving channels would be the

following:

Transmitted Laser Beams(1,550 and 980 nm)

Transmitted Beam(1,550 nm)

Single-Mode Optical Fiber

Transmitted Reference Beam (980 nm)Returning From Corner-Cube Reflector

Beacon Beam FromGround Station (852 nm)

External Corner-CubeReflector (Not Part of

Optical Assembly)

Coarse-Acquisition and TransmittingReference Channels

Fine-Acquisition, Tracking, andTransmitting Reference Channels

The Optical Assembly of the OCT would be compact, yet would accommodate six optical channels, each playing a different role in transmission or reception.


• A channel for a portion of an 852-nm-wavelength beacon-and-data-commu-nication signal from a ground stationfor use in the coarse-acquisition con-trol system;

• A channel for another portion of the852-nm signal for use in the fine-acqui-sition-and-tracking system; and

• A channel for yet another portion ofthe 852-nm signal, used for receptionof data from the ground station.The telescope in the OCT would have

an aperture 100 mm wide and would beafocal: all beams would be collimated atthe points where they would be split. Thedesign would minimize vignetting andwould include field stops, Lyot stops, andbaffles to block stray light. To make theoptical system compact, the primary mir-ror would have a focal-length/diameterratio (“f” number) of 1.2.

In the first-mentioned transmittingchannel, the 1,550-nm laser light com-ing from a single-mode optical fiberwould be collimated and directed to aspot on the FSM coincident with a pupilimage plane, then reflected from theFSM to a dichroic beam splitter (DBS),then reflected by four more mirrors,the last two of which would be the sec-ondary and primary telescope mirrors.The divergence of the outgoing 1,550-nm laser beam could be tailored by al-tering the design of the collimating in-terface module: one would choose theamount of divergence according to

range of the free-space optical link andthe degree of mechanical stability ofthe aircraft to carry the OCT. The FSMcould steer the 1,550-nm laser beamover an angular range about 10 millira-dians wide.

In the second-mentioned transmittingchannel (the one used for reference forfine pointing), the 980-nm laser beamwould be made to propagate with the1,550-nm beam through the single-mode optical fiber and the rest of theoptical train until the two beams reachthe DBS. At the DBS, a significant frac-tion of the 980-nm beam would be trans-mitted through relay optics to a retrore-flector. The retroreflected 980-nm beamwould be guided back to a second beamsplitter, where the reflected fractionwould be brought to focus at a focalplane (the fine-acquisition focal plane),the field of view of which would be 10milliradians wide. Thus, steering by theFSM would change the location of the980-nm-wavelength beam spot on thisfocal plane.

The fraction of the 980-nm beamtransmitted through the DBS wouldpropagate through the rest of the opti-cal train and out of the telescope alongwith the 1,550-nm beam. If the telescopewere to be deliberately mechanicallyaimed at an external corner-cube reflec-tor, the reflected portion of the 980-nmwould travel back into the telescope,through the DBS, and onto the fine-ac-

quisition focal plane to a spot differentfrom the reference spot mentioned inthe previous paragraph.

Another portion of the returning 980-nm beam, constituting the beam in thethird-mentioned transmitting channel,would impinge on the coarse-acquisitionfocal plane (that is, on the coarse-acqui-sition CCD). This arrangement would fa-cilitate the calibration of co-boresighted-ness between the coarse-acquisition andfine-steering fields of view.

In the first-mentioned receiving chan-nel, a portion of the 852-nm signal fromthe ground station would impinge onthe coarse-acquisition focal plane, whichwould have a field of view 3° wide —wide enough to facilitate acquisitionunder most circumstances. In the sec-ond-mentioned receiving channel, aportion of the 852-nm signal would beguided to a focus on the fine-acquisitionfocal plane. The design would ensurethat location of this focus would differfrom that of the 980-nm beam. In thethird-mentioned receiving channel, thebeam splitter would divert a fraction ofthe received 852-nm beam to a detectorthat would extract any data signal con-veyed as modulation of this beam.

This work was done by Abhijit Biswas,Norman Page, and Hamid Hemmati of Cal-tech for NASA’s Jet Propulsion Labora-tory. Further information is contained in aTSP (see page 1).NPO-30537


An algorithm analyzes rain-gauge datato identify statistical outliers that could bedeemed to be erroneous readings.Heretofore, analyses of this type havebeen performed in burdensome manualprocedures that have involved subjectivejudgements. Sometimes, the analyses haveincluded computational assistance for de-tecting values falling outside of arbitrarylimits. The analyses have been performedwithout statistically valid knowledge of thespatial and temporal variations of precipi-tation within rain events. In contrast, thepresent algorithm makes it possible to au-tomate such an analysis, makes the analy-sis objective, takes account of the spatialdistribution of rain gauges in conjunctionwith the statistical nature of spatial varia-tions in rainfall readings, and minimizesthe use of arbitrary criteria.

The algorithm implements an itera-tive process that involves nonparamet-ric statistics. The steps of the algorithmare the following:1. Raw rain-gauge data are subjected to

qualitative tests of validity. The detailsof the tests are attuned to the detailsof the sources of data and data-entryprocedures. For example, reports thatinclude negative rain-gauge readingsor incorrect dates are rejected. Datathat pass these tests are accepted forprocessing in the next step.

2. Associated with each gauge is aneighborhood, defined as that gaugeplus the five nearest gauges that (a)

have reported, (b) are currently ac-cepted, and (c) are more than 100meters distant.The 100-meter distance criterion is ar-

bitrary, but not totally so: It has beenchosen to ensure that each acceptedgauge gives a reading independent ofthat of any other accepted gauge. Inde-pendence of readings is basic assump-tion of the statistical analysis performedin the subsequent steps.

The five-nearest-gauge criterion isalso only partly arbitrary: It has beenchosen as a compromise between (a)undesired sensitivity to numerical arti-facts at fewer gauges per neighborhoodand (b) undesired insensitivity to inputerrors (which the errors that one seeksto detect) at greater numbers of gaugesper neighborhood.3. The six readings from each neigh-

borhood are ranked. If the readingof the gauge under consideration is alocal minimum or maximum, then itis deemed erroneous if it is less thanone-third or greater than three timesthe reading of the gauge of the adja-cent rank.

4. After rejection of the gauges that havebeen thus deemed to give erroneousreadings, a new set of neighborhoodsis computed from the remaining ac-cepted gauges, again following thelogic of step 2.

5. The readings from gauges in thenew neighborhoods are examined

for errors, again following the logicof step 3.

6. The neighborhood of any gauge instep 3 or step 5 is examined to deter-mine which, if any, other gauges in theneighborhood also were flagged as giv-ing erroneous readings. If all of thegauges in the neighborhood havebeen flagged and if, in addition, theirerrors have all been found to be of thesame sense (that is, all high or all low),then the readings from the neighbor-hood are assumed to be correct. Thejustification for this decision is that it isunlikely that two or more independ-ent, spatially adjacent observationswould both be extreme highs or ex-treme lows. In addition, when a gaugeis flagged because of a low reading andthe readings of at least three othergaugesare zero but are not local min-ima, then that gauge is not flagged.The algorithm has been imple-

mented as a series of subroutines in acomputer program used to edit sets ofrainfall data. The algorithm could alsobe implemented as a program in itsown right or incorporated into otherprograms for the purpose of identifyingerroneous input data pertaining to phe-nomena other than rainfall.

This work was done by Doug Rickman ofMarshall Space Flight Center. Furtherinformation is contained in a TSP (seepage 1).MFS-31993-1

Information Sciences

Algorithm for Identifying Erroneous Rain-Gauge ReadingsWhat was previously a subjective manual analysis is automated and made objective.Marshall Space Flight Center, Alabama

Condition Assessment and End-of-Life Prediction System forElectric Machines and Their LoadsSystem generates on-line, real-time condition assessment and end-of-life prediction.Lyndon B. Johnson Space Center, Houston, Texas

An end-of-life prediction system de-veloped for electric machines and theirloads could be used in integrated vehi-cle health monitoring at NASA and inother government agencies. This systemwill provide on-line, real-time conditionassessment and end-of-life prediction of

electric machines (e.g., motors, genera-tors) and/or their loads of mechanicallycoupled machinery (e.g., pumps, fans,compressors, turbines, conveyor belts,magnetic levitation trains, and others).In long-duration space flight, the abilityto predict the lifetime of machinery

could spell the difference between mis-sion success or failure. Therefore, thesystem described here may be of ines-timable value to the U.S. space pro-gram.

No known system (hardware, soft-ware, or hybrids) currently exists that


performs this function. While there areseveral software programs used com-mercially that address various aspects ofthe off-line diagnostics analysis of rotat-ing equipment — including motors,generators, turbines, and the like — theineffectiveness of these programs in di-agnosing incipient failures is well-docu-mented. Indeed, industry feedback sug-gests that the majority of these softwaretools will provide accurate diagnoses inonly about 60 to 65 percent of analyzedcases.

The system differs greatly from com-mercial software because the systemwill provide continuous monitoring foron-line condition assessment and end-of-life prediction as opposed to the cur-rent off-line diagnoses. Similarly, thesystem will also provide real-time con-dition assessment and end-of-life pre-diction as contrasted with the delayeddiagnosis of commercially available

software. Indeed, this system also willprovide automated condition assess-ment and end-of-life prediction (notmanual fault diagnosis analysis), inte-grated condition assessment and end-of-life prediction capability (not thecurrent sensor-dependent fault diagno-sis analysis), and machine- and/orload-independent condition assess-ment and end-of-life prediction, whichdiffers significantly from the machine-dependent diagnosis of commercialsoftware systems. In sum, therefore,this invention provides the followingnew and critical features:• an all-in-one on-line, real-time condi-

tion assessment and end-of-life predic-tion system for electric machines andtheir loads;

• a machine- and load-independent con-dition assessment and end-of-life pre-diction system; and

• enhanced effectiveness resulting from

the information-processing technolo-gies used. The system clearly advances the state-

of-the-art. Its software is independent ofany specific hardware platform, nothird-party programs are required foroperations, and the system can be imple-mented with any high- or low-level pro-gramming language. Potential users ofthis software will run the gamut from thespace program (for which it was devel-oped) to anyone who maintains electricmachines and their loads.

This work was done by Alexander G. Parlosand Hamid A. Toliyat of Texas A&M Univer-sity, Departments of Mechanical and ElectricalEngineering for Johnson Space Center. Forfurther information, contact:

Alexander G. ParlosE-mail: [email protected]

Refer to MSC-22894, volume and number of thisNASA Tech Briefs issue, and the page number.


Books & Reports

Lightweight Thermal Insula-tion for a Liquid-OxygenTank

A proposed lightweight, reusable ther-mal-insulation blanket has been designedfor application to a tank containing liquidoxygen, in place of a non-reusable spray-on insulating foam. The blanket would beof the multilayer-insulation (MLI) typeand equipped with a pressure-regulatednitrogen purge system. The blanket wouldcontain 16 layers in two 8-layer sub-blan-kets. Double-aluminized polyimide 0.3 mil(≈0.008 mm) thick was selected as a reflec-tive shield material because of its compati-bility with oxygen and its ability to with-stand ionizing radiation and hightemperature. The inner and outer sub-blanket layers, 1 mil (≈0.025 mm) and 3mils (≈0.076 mm) thick, respectively,would be made of the double-aluminizedpolyimide reinforced with aramid. Theinner and outer layers would providestructural support for the more fragile lay-ers between them and would bear the in-sulation-to-tank attachment loads. The lay-ers would be spaced apart by lightweight,low-thermal-conductance netting madefrom polyethylene terephthalate.

This work was done by G. Scott Willen, Jen-nifer Lock, and Steve Nieczkoski of TechnologyApplications, Inc. for Johnson Space Center.For further information, contact:

Technology Applications, Inc.5445 Conestoga Court, #2ABoulder, CO 80301-2724Telephone No.: (303) 443-2262www.techapps.com.

Refer to MSC-23099.

Stellar Gyroscope for Deter-mining Attitude of a Spacecraft

A paper introduces the concept of a stel-lar gyroscope, currently at an early stage ofdevelopment, for determining the attitudeor spin axis, and spin rate of a spacecraft.Like star trackers, which are commerciallyavailable, a stellar gyroscope would captureand process images of stars to determinethe orientation of a spacecraft in celestialcoordinates. Star trackers utilize charge-coupled devices as image detectors and arecapable of tracking attitudes at spin rates ofno more than a few degrees per second andupdate rates typically <5 Hz. In contrast, astellar gyroscope would utilize an active-pixel sensor as an image detector andwould be capable of tracking attitude at aslew rate as high as 50°/s, with an updaterate as high as 200 Hz. Moreover, a stellargyroscope would be capable of measuring aslew rate up to 420°/s. Whereas a Sun sen-sor and a three-axis mechanical gyroscopeare typically needed to complement a startracker, a stellar gyroscope would functionwithout them; consequently, the mass,power consumption, and mechanical com-plexity of an attitude-determination systemcould be reduced considerably.

This work was done by Bedabrata Pain,Bruce Hancock, Carl Liebe, and Jeffrey Mell-strom of Caltech for NASA’s Jet PropulsionLaboratory. Further information is con-tained in a TSP (see page 1). NPO-30481

Lifting Mechanism for theMars Explorer Rover

A report discusses the design of a roverlift mechanism (RLM) — a major subsys-tem of each of the Mars Exploration Rovervehicles, which were landed on Mars inJanuary 2004. The RLM had to satisfy re-quirements to (1) be foldable as part of anextremely dense packing arrangementand (2) be capable of unfolding itself in acomplex, multistep process for disengag-ing the rover from its restraints in the lan-der, lifting the main body of the rover offits landing platform, and placing the roverwheels on the platform in preparation fordriving the rover off the platform. Therewas also an overriding requirement tominimize the overall mass of the rover andlander. To satisfy the combination of theseand other requirements, it was necessaryto formulate an extremely complex designthat integrated components and functionsof the RLM with those of a rocker-bogiesuspension system, the aspects of whichhave been described in several prior NASATech Briefs articles. In this design, suspen-sion components also serve as parts of a 4-bar linkage in the RLM.

This work was done by Joseph Melko,Theodore Iskenderian, Brian Harrington, andChristopher Voorhees of Caltech for NASA’s JetPropulsion Laboratory. Further informationis contained in a TSP (see page 1).NPO-40875

National Aeronautics andSpace Administration

technology focus computers/electronics

Documents