Technology Focus Electronics/Computers · 2013-04-10 · NASA Tech Briefs, November 2008 1 INTRODUCTION Tech Briefs are short announcements of innovations originating from research

Technology Focus

Electronics/Computers

Software

Materials

Mechanics/Machinery

Manufacturing

Bio-Medical

Physical Sciences

Information Sciences

Books and Reports

11-08 November 2008

https://ntrs.nasa.gov/search.jsp?R=20080047178 2020-03-15T01:55:55+00:00Z

NASA Tech Briefs, November 2008 1

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

ment activities of the National Aeronautics and Space Administration. They emphasize information 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 http://www.nttc.edu/about/contactus.asp

Please reference the control numbers appearing at the end of each Tech Brief. Infor mation 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 listed below.

Ames Research CenterLisa L. Lockyer(650) [email protected]

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

Glenn Research CenterKathy Needham(216) [email protected]

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

Jet Propulsion LaboratoryKen Wolfenbarger(818) [email protected]

Johnson Space CenterMichele Brekke(281) [email protected]

Kennedy Space CenterDavid R. Makufka(321) [email protected]

Langley Research CenterMartin Waszak(757) [email protected]

Marshall Space Flight CenterJim Dowdy(256) [email protected]

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

Carl Ray, Program ExecutiveSmall Business Innovation Research (SBIR) & Small Business Technology Transfer (STTR) Programs(202) [email protected]

Doug Comstock, DirectorInnovative Partnerships Program Office(202) [email protected]

NASA Field Centers and Program Offices

5 Technology Focus: Test & Measurements

5 Digital Phase Meter for a Laser HeterodyneInterferometer

6 Vision System Measures Motions of Robot andExternal Objects

6 Advanced Precipitation Radar Antenna ToMeasure Rainfall From Space

7 Wide-Band Radar for Measuring Thickness of Sea Ice

8 Vertical Isolation for Photodiodes in CMOS Imagers

9 Wide-Band Microwave Receivers Using Photonic Processing

11 Electronics/Computers11 L-Band Transmit/Receive Module for Phase-Stable

Array Antennas 11 Microwave Power Combiner/Switch Utilizing a

Faraday Rotator12 Compact Low-Loss Planar Magic-T12 Using Pipelined XNOR Logic to Reduce SEU Risks

in State Machines13 Quasi-Optical Transmission Line for 94-GHz Radar13 Next Generation Flight Controller Trainer System

15 Software15 Converting From DDOR SASF to APF15 Converting From CVF to AAF15 Documenting AUTOGEN and APGEN Model Files15 Sequence History Update Tool15 Extraction and Analysis of Display Data16 MRO DKF Post-Processing Tool16 Rig Diagnostic Tools16 MRO Sequence Checking Tool16 Science Activity Planner for the MER Mission 16 UAVSAR Flight-Planning System

19 Manufacturing & Prototyping19 Templates for Deposition of Microscopic Pointed

Structures

21 Materials21 Adjustable Membrane Mirrors Incorporating G-

Elastomers21 Hall-Effect Thruster Utilizing Bismuth as

Propellant21 High-Temperature Crystal-Growth Cartridge

Tubes Made by VPS 22 Quench Crucibles Reinforced With Metal

23 Mechanics/Machinery23 Deep-Sea Hydrothermal-Vent Sampler24 Mars Rocket Propulsion System24 Two-Stage Passive Vibration Isolator24 Improved Thermal Design of a Compression Mold25 Enhanced Pseudo-Waypoint Guidance for

Spacecraft Maneuvers

27 Physical Science27 Altimetry Using GPS-Reflection/Occultation

Interferometry27 Thermally Driven Josephson Effect27 Perturbation Effects on a Supercritical C7H16/N2

Mixing Layer28 Gold Nanoparticle Labels Amplify Ellipsometric

Signals28 Phase Matching of Diverse Modes in a WGM

Resonator29 WGM Resonators for Terahertz-to-Optical

Frequency Conversion30 Determining Concentration of Nanoparticles

From Ellipsometry31 Microwave-to-Optical Conversion in WGM

Resonators31 Four-Pass Coupler for Laser-Diode-Pumped Solid-

State Laser32 Low-Resolution Raman-Spectroscopy Combustion

Thermometry 34 Temperature Sensors Based on WGM Optical

Resonators35 Varying the Divergence of Multiple Parallel Laser

Beams

37 Information Sciences37 Efficient Algorithm for Rectangular Spiral Search37 Algorithm-Based Fault Tolerance Integrated

With Replication38 Targeting and Localization for Mars Rover

Operations38 Terrain-Adaptive Navigation Architecture 38 Self-Adjusting Hash Tables for Embedded Flight

Applications39 Schema for Spacecraft-Command Dictionary

41 Books & Reports41 Combined GMSK Communications and PN

Ranging 41 System-Level Integration of Mass Memory41 Network-Attached Solid-State Recorder

Architecture

41 Method of Cross-Linking Aerogels Using a One-Pot Reaction Scheme

41 An Efficient Reachability Analysis Algorithm

11-08 November 2008


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


Technology Focus: Test & Measurements

Digital Phase Meter for a Laser Heterodyne InterferometerThe design is suitable for numerous commercial products that utilize phase measurements.NASA’s Jet Propulsion Laboratory, Pasadena, California

The figure depicts two digital wave-forms and a block diagram of a digitalphase meter for measuring the differ-ence between their phases. This digitalphase meter is being developed for in-corporation into a laser heterodyne in-terferometer in a metrological appara-tus, but could also be adapted to otheruses. Relative to prior phase meters ofsimilar capability, including digital ones,this digital phase meter is smaller, lesscomplex, and less expensive. The phasemeter has been constructed and testedin the form of a field-programmable gatearray (FPGA) of fewer than 104 gates.

A description of the signals to beprocessed is prerequisite to a meaningfuldescription of this digital phase meter.There are two digital, half-duty-cyclesquare-wave input signals, denoted the ref-erence and metrological input, respec-tively. Both signals oscillate at the same fre-quency, which, in the original intendedapplication, is the laser heterodyne fre-quency (typically of the order of 100 kHz).In the original intended application, thephase difference between the two signals isdirectly proportional to a relative changein the lengths of portions of an opticalpath traversed by the laser beam.

The three main components at theheart of the phase meter are a phase de-tector, a correction clock-signal genera-tor, and a digital phase shifter. Thesecomponents are connected to form aphase-locked loop (PLL) that is aslightly modified version of a conven-tional PLL. The modification consistsmainly in feeding the metrological inputsignal, instead of the reference input sig-nal, to one of the input terminals of thephase shifter. That is to say, whereas thephase of the reference signal gets shiftedin a conventional PLL, the phase of themetrological signal is the one that getsshifted in this PLL.

The reference signal is fed to one ofthe two inputs of a phase detector, whilethe phase-shifted metrological signal isfed to the other input of the phase de-tector, which measures the differencebetween the phases of its inputs and putsout a correction command (basically, a

pair of error feedback signals) in an ef-fort to drive the difference toward zero.The correction command is fed to thecorrection clock-signal generator, theoutputs of which are fed as inputs to thedigital phase shifter. The overall actionof the loop is to shift the phase of themetrological signal until it becomesphase-aligned with the reference signal.When this alignment is achieved, thephase detector stops sending correctioncommands to the phase shifter.

One of the outputs of the phase detec-tor is a number proportional to the in-

stantaneous phase shift imposed on themetrological signal. This number is fedto an averaging generator and a registerset interface for display of instantaneousand time-averaged phase readings, whichcan include readings from which initialvalues have been subtracted so as to indi-cate the change in phase shift from theinitial to the current or final state of ametrological process.

This work was done by Frank Loya of Cal-tech for NASA’s Jet Propulsion Laboratory. Fur-ther information is contained in a TSP (seepage 1). NPO-40318

PhaseDifference

ReferenceInput

MetrologicalInput

Inst

anta

neo

us

Am

plit

ud

e

Time or PhaseINPUT SIGNALS

PHASE METER

Input Metrological Signal

Phase-Shifted Metrological Signal

InputReferenceSignal

PhaseDetector

CorrectionClock-SignalGenerator

ErrorDetected

ErrorDirection

Metrological-Wave-TableWrite-Up Count

Metrological-Wave-TableRead-Up Count

EnableMetrological-Wave-TableRead-Up Count

DigitalPhase Shifter

(Volatge-ControlledOscillator)

Write-Time, Read-Time,Shift-Time, andQuiet-Time PulsesSystem

Clock-SignalsGenerator

Averaging Generator

Register Set Interface(for Instantaneous and Average Count Readout)

Start StopBeginning

CountSample

CountDifference,

Sum

InstantaneousShift Count

The Digital Phase Meter is based on a modified phase-locked loop. When phase alignment betweenthe reference input and the phase-shifted metrological input is achieved, the loop locks and the phaseshift of the digital phase shifter equals the phase difference that one seeks to measure.

6 NASA Tech Briefs, November 2008

Advanced Precipitation Radar Antenna To Measure Rainfall From Space This parabolic cylindrical reflector uses Ku and Ka bands. NASA’s Jet Propulsion Laboratory, Pasadena, California

To support NASA’s planned 20-yearmission to provide sustained global pre-cipitation measurement (EOS-9 GlobalPrecipitation Measurement (GPM)), adeployable antenna has been exploredwith an inflatable thin-membrane struc-ture. This design uses a 5.3×5.3-m inflat-able parabolic reflector with the electron-ically scanned, dual-frequency phasedarray feeds to provide improved rainfallmeasurements at 2.0-km horizontal reso-lution over a cross-track scan range of upto ±37º, necessary for resolving intense,

isolated storm cells and for reducing thebeam-filling and spatial sampling errors.The two matched radar beams at the twofrequencies (Ku and Ka bands) will allowunambiguous retrieval of the parametersin raindrop size distribution.

The antenna is inflatable, usingrigidizable booms, deployable chain-link supports with prescribed curva-tures, a smooth, thin-membrane reflect-ing surface, and an offset feedtechnique to achieve the precision sur-face tolerance (0.2 mm RMS) for meet-

ing the low-sidelobe requirement. Thecylindrical parabolic offset-feed reflec-tor augmented with two linear phasedarray feeds achieves dual-frequencyshared-aperture with wide-angle beamscanning and very low sidelobe level of–30 dB. Very long Ku and Ka band mi-crostrip feed arrays incorporating acombination of parallel and seriespower divider lines with cosine-over-pedestal distribution also augment thesidelobe level and beam scan. This de-sign reduces antenna mass and launch

A prototype of an advanced robotic vi-sion system both (1) measures its ownmotion with respect to a stationary back-ground and (2) detects other movingobjects and estimates their motions, allby use of visual cues. Like some prior ro-botic and other optoelectronic visionsystems, this system is based partly onconcepts of optical flow and visualodometry. Whereas prior optoelectronicvisual-odometry systems have been lim-ited to frame rates of no more than 1 Hz,a visual-odometry subsystem that is partof this system operates at a frame rate of60 to 200 Hz, given optical-flow esti-mates. The overall system operates at aneffective frame rate of 12 Hz. Moreover,unlike prior machine-vision systems fordetecting motions of external objects,this system need not remain stationary:it can detect such motions while it ismoving (even vibrating).

The system includes a stereoscopic pairof cameras mounted on a moving robot.The outputs of the cameras are digitized,then processed to extract positions andvelocities. The initial image-data-process-ing functions of this system are the sameas those of some prior systems: Stere-oscopy is used to compute three-dimen-sional (3D) positions for all pixels in thecamera images. For each pixel of eachimage, optical flow between successiveimage frames is used to compute the two-

dimensional (2D) apparent relative trans-lational motion of the point transverse tothe line of sight of the camera.

The challenge in designing this sys-tem was to provide for utilization of the3D information from stereoscopy in con-junction with the 2D information fromoptical flow to distinguish between mo-tion of the camera pair and motions ofexternal objects, compute the motion ofthe camera pair in all six degrees oftranslational and rotational freedom,and robustly estimate the motions of ex-ternal objects, all in real time. To meetthis challenge, the system is designed toperform the following image-data-pro-cessing functions:

The visual-odometry subsystem (thesubsystem that estimates the motion ofthe camera pair relative to the stationarybackground) utilizes the 3D informationfrom stereoscopy and the 2D informa-tion from optical flow. It computes therelationship between the 3D and 2D mo-tions and uses a least-mean-squares tech-nique to estimate motion parameters.The least-mean-squares technique issuitable for real-time implementationwhen the number of external-moving-object pixels is smaller than the numberof stationary-background pixels.

In another subsystem, pixels represen-tative of external transversely moving ob-jects are detected by means of differ-

ences between (1) apparent transversevelocities computed from optical flowand (2) the corresponding relativetransverse velocities estimated from vi-sual odometry under the temporary as-sumption that all pixels belong to thestationary background.

In yet another subsystem, pixels repre-sentative of radially moving objects aredetected by means of differences be-tween (1) changes in radial distance esti-mated from changes in stereoscopic dis-parities between successive image framesand (2) the corresponding relative ra-dial velocities estimated from visualodometry under the temporary assump-tion that all pixels belong to the station-ary background. However, it is more dif-ficult to detect radial than to detecttransverse motion, especially at large dis-tances. This difficulty is addressed by in-corporating several additional process-ing features, including means toestimate rates of change of stereoscopicdisparities, post-processing to preventfalse alarms at low signal-to-noise ratios,and taking advantage of sometimesbeing able to distinguish between radial-motion optical flow and transverse-mo-tion optical flow at short distances.

This work was done by Ashit Talukder andLarry Matthies of Caltech for NASA’s JetPropulsion Laboratory. Further information iscontained in a TSP (see page 1). NPO-40687

Vision System Measures Motions of Robot and External ObjectsFrame rates greatly exceed those of prior systems. NASA’s Jet Propulsion Laboratory, Pasadena, California


vehicle stowage volume. The Ku and Kaband feed arrays are needed to achievethe required cross-track beam scan-ning. To demonstrate the inflatablecylindrical reflector with two linear po-larizations (V and H), and two beam di-rections (0º and 30º), each frequencyband has four individual microstrip

array designs. The Ku-band array has atotal of 166×2 elements and the Ka-band has 166×4 elements with bothbands having element spacing about0.65 λ0.

The cylindrical reflector with offsetlinear array feeds reduces the complex-ity from “N×N” transmit/receive (T/R)

modules of a conventional planar-phased array to just “N” T/R modules.The antenna uses T/R modules withelectronic phase-shifters for beam steer-ing. The offset reflector does not pro-vide poor cross-polarization like a dou-ble-curved offset reflector would, and itallows the wide scan angle in one planerequired by the mission. Also, the cylin-drical reflector with two linear arrayfeeds provides dual-frequency perform-ance with a single, shared aperture. Theaperture comprises a reflective surfacewith a focal length of 1.89 m and is madefrom aluminized Kapton film. The re-flective surface is of uniform thickness inthe range of a few thousandths of aninch and is attached to the chain-linksupport structure via an adjustable sus-pension system. The film aperture rollsup, together with the chain-link struc-ture, for launch and can be deployed inspace by the deployment of the chain-link structure.

This work was done by Yahya Rahmat-Samii of UCLA; John Lin of ILC Dover, Inc.;and John Huang, Eastwood Im, MichaelLou, Bernardo Lopez, and Stephen Durdenof Caltech for NASA’s Jet Propulsion Labora-tory. For more information, contact [email protected]. NPO-40470

Mandrel

InflatableRigidizable Rib

InflatableRigidizable

Precision“Chain” Linkage

SpacecraftBus

Feed Arrays

The configuration of the radar antenna features a Chain-Link Support Structure that is space-de-ployable.

A wide-band penetrating radar sys-tem for measuring the thickness of seaice is under development. The needfor this or a similar system arises as fol-lows: Spatial and temporal variations inthe thickness of sea ice are importantindicators of heat fluxes between theocean and atmosphere and, hence, areimportant indicators of climate changein polar regions. A remote-sensing sys-tem that could directly measure thethickness of sea ice over a wide thick-ness range from aboard an aircraft orsatellite would be of great scientificvalue. Obtaining thickness measure-ments over a wide region at weekly ormonthly time intervals would con-tribute significantly to understandingof changes in the spatial distributionand of the mass balance of sea ice.

A prototype of the system was de-signed on the basis of computationalsimulations directed toward under-standing what signal frequencies are

needed to satisfy partly competing re-quirements to detect both bottom andtop ice surfaces, obtain adequate pene-tration despite high attenuation in thelossy sea-ice medium, and obtain ade-quate resolution, all over a wide thick-ness range. The prototype of the sys-tem is of the frequency-modulation,continuous-wave (FM-CW) type. At agiven time, the prototype functions ineither of two frequency-band/operational-mode combinationsthat correspond to two thicknessranges: a lower-frequency (50 to 250MHz) mode for measuring thicknessgreater than about 1 m, and a higher-frequency (300 to 1,300 MHz) modefor measuring thickness less than about1 m. The bandwidth in the higher-fre-quency (lesser-thickness) mode is ade-quate for a thickness resolution of 15cm; the bandwidth in the lower-fre-quency (greater-thickness) mode is ad-equate for a thickness resolution of 75

cm. Although a thickness resolution ofno more than 25 cm is desired for sci-entific purposes, the 75-cm resolutionwas deemed acceptable for the purposeof demonstrating feasibility.

The prototype was constructed as amodified version of a 500-to-2,000-MHzFM-CW radar system developed previ-ously for mapping near-surface internallayers of the Greenland ice sheet. Theprototype included two sets of antennas:one for each frequency-band/mode. ForArctic and Antarctic field tests, the pro-totype was mounted on a sled that wastowed across the ice. The Arctic field testwas performed in the lower-frequencymode on ice ranging in thickness from 1to 4 m. In the analysis of the results ofthe Arctic field test, a comparison of theradar-determined ice thicknesses withactual ice thicknesses yielded an overallmean difference of 14 cm and standarddeviation of 30 cm. The Antarctic fieldtest was performed in the higher-fre-

Wide-Band Radar for Measuring Thickness of Sea Ice This instrument could contribute to understanding of climate change. NASA’s Jet Propulsion Laboratory, Pasadena, California


quency mode; analysis of the results ledto the conclusion that this mode is use-ful for measuring thicknesses between0.5 and 1 m.

Several modifications have been con-ceived for implementation in further de-velopment toward an improved practicalsystem: • The system would function in a single

frequency-band/mode (100 to 1,200MHz) that would afford a resolution ofabout 15 cm.

• There would be a single antenna sys-tem that would be optimized for theentire 100-to-1,000-MHz frequencyband.

• To enable ice-thickness surveys overlarger areas, the system would be madecapable of operating aboard a low-fly-ing aircraft that could be either pilotedor robotic.

• Data-processing techniques to decon-volve the system response have beendeveloped on the basis of impulse-re-

sponse measurements over a calmocean. Implementation of these tech-niques in the system would enable cor-rection for imperfections of the systemand would thereby increase the effec-tive sensitivity of the system. This work was done by Prasad Gogineni

and Pannir Kanagaratnam of the Univer-sity of Kansas and Benjamin M. Holt ofCaltech for NASA’s Jet Propulsion Labora-tory. Further information is contained in aTSP (see page 1). NPO-45565

Vertical Isolation for Photodiodes in CMOS Imagers Diffusion cross-talk would be reduced substantially. NASA’s Jet Propulsion Laboratory, Pasadena, California

In a proposed improvement in com-plementary metal oxide/semi con duct -or (CMOS) image detectors, two addi-tional implants in each pixel wouldeffect vertical isolation between themetal oxide/semiconductor field-effecttransistors (MOSFETs) and the photo-diode of the pixel. This improvement isexpected to enable separate optimiza-tion of the designs of the photodiodeand the MOSFETs so as to optimizetheir performances independently ofeach other. The purpose to be served byenabling this separate optimization isto eliminate or vastly reduce diffusioncross-talk, thereby increasing sensitivity,effective spatial resolution, and color fi-delity while reducing noise.

Ideally, the spatial resolution of animager should be limited by the geo-metric pixel size. However, in mostpractical image detectors, resolutionsare limited, not by geometric pixelsizes, but by cross-talk. (As used here,“cross-talk” denotes the response of apixel to light focused on an adjacentpixel.) Cross-talk degrades spatial reso-lution of an imager, reduces overallsensitivity, compromises color fidelity,and leads to additional noise in theimage after color correction. Diffusioncross-talk occurs where photogener-ated charge carriers can move to neigh-boring charge-accumulation sites — inparticular, where junction diodes in ad-jacent pixels have insufficient deple-tion widths.

The left side of the figure presents aschematic cross section of a typical con-ventional CMOS imager pixel contain-ing a junction diode connected to thesource of a reset MOSFET. The junc-tion diode is formed between the n

well and the p epitaxial layer (or p sub-strate). The n well is connected to thesource of the reset MOSFET throughan n+ implant. The reset MOSFET andan associate source-follower MOSFETare n-type and are placed inside a pwell. For reasons too complex to pres-ent in this article, the depletion widthis too small to prevent lateral diffusionof photo-induced charge carriers in the undepleted (field-free) epitaxialregion. In the absence of a guidingelectric field, photoelectrons gener-ated in the epitaxial layer substrate dif-fuse omnidirectionally between pixels,thereby causing cross-talk.

The maximum supply potential in aCMOS process is between 3 and 5 V.When potential drops are taken into ac-count, the reverse bias across the diodeis between 2 and 3 V. At these reversebiases, the p-n junction depletion widthis too small to prevent diffusion cross-

talk, especially for longer wavelengthlight, In principle, the depletion widthcould be increased significantly by ap-plying a large reverse bias (e.g., 50 V) tothe p epitaxial layer or substrate. How-ever, because of (1) the electrical con-nection between the p well and the pepitaxial layer or substrate and (2) a re-quirement to keep at the most between3 and 5 V across the CMOS devices, it isnot possible to apply such a large re-verse bias in this device structure. Thisprompts the proposed improvement indevice structure.

A CMOS imager pixel as proposed,depicted on the right side of the fig-ure, would include a deep n well and adeep p well in addition to the conven-tional n and p wells. The photodiodewould be formed by the deep n welland the p epitaxial layer or substrate.The anode end (n end) of the diodewould be connected to the n+ source

The Proposed CMOS Imager Pixel device structure would be similar to the conventional one but wouldinclude deep p and n wells.

Deep n Well

p Wellp Well

p E

pit

axia

l Lay

er o

r Su

bst

rate

n Welln Welln Well

FieldOxide

FieldOxide

FieldOxide

n+Implant

Deep p Well

Depletion Region

DepletionRegion

DepletionRegion

Connecting Metal Connecting MetalGate

n+Implant

p Ep

itaxial Layer or Su

bstrate

CONVENTIONAL PROPOSED

Passivation

/Isolatio

n Layer


implant of the reset MOSFET throughthe conventional n well. The reset and source-follower MOSFETs wouldreside in the p well as in the conven-tional device structure.

Unlike in the conventional devicestructure, the deep n well would elec-trostatically separate the p well in thevertical direction from the p epitaxiallayer or substrate. The horizontal isola-tion of photodiodes in adjacent pixelsfrom each other would be achieved bythe deep p wells: Each deep p wellwould establish a potential barrier thatwould prevent electrons in the deep nwells of adjacent pixels from communi-cating with each other.

Inasmuch as the conventional anddeep p wells would both be electrosta-

tically isolated from the p epitaxiallayer or substrate by the deep n well,any reverse (negative) bias could beapplied to the p epitaxial layer or sub-strate without causing the potentialdifference between the n and p wellsto increase beyond the typical conven-tional range of 2 to 3 V. Dependingupon the resistivity of the substrate, aback-side reverse bias in excess of 50 Vcould be applied to achieve depletionwidths as large as 50 µm, while theMOSFETs could be operated with con-ventional CMOS power supplies andbiases. Thus, the incorporation of thedeep n well and p well would allow theintegration of a photodiode with a verylarge back-bias and very large deple-tion width alongside state-of-the-art

MOSFETs with small supply voltages,resulting in the development of high-performance CMOS imager sensors.

This work was done by Bedabrata Pain ofCaltech for NASA’s Jet Propulsion Laboratory.

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-41226, volume and number

of this NASA Tech Briefs issue, and thepage number.

Wide-Band Microwave Receivers Using Photonic ProcessingOne receiver would have the functionality of multiple traditional heterodyne microwave receivers.NASA’s Jet Propulsion Laboratory, Pasadena, California

In wide-band microwave receivers ofa type now undergoing development,the incoming microwave signals areelectronically preamplified, then fre-quency-up-converted to optical signalsthat are processed photonically beforebeing detected. This approach differsfrom the traditional approach, inwhich incoming microwave signals areprocessed by purely electronic means.As used here, “wide-band microwavereceivers” refers especially to receiverscapable of reception at any frequencythroughout the range from about 90 toabout 300 GHz. The advantage ex-pected to be gained by following theup-conversion-and-photonic-process-ing approach is the ability to overcomethe limitations of currently availabledetectors and tunable local oscillatorsin the frequency range of interest.

In a receiver following this approach(see figure), a preamplified incomingmicrowave signal is up-converted by the method described in the preceed-ing article. The frequency up-converterexploits the nonlinearity of the electro-magnetic response of a whispering-gallery-mode (WGM) resonator madeof LiNbO3. Up-conversion takes placeby three-wave mixing in the resonator.The WGM resonator is designed andfabricated to function simultaneously

as an electro-optical modulator and toexhibit resonance at the microwaveand optical operating frequencies plusphase matching among the microwaveand optical signals circulating in theresonator. The up-conversion is an effi-cient process, and the efficiency is en-hanced by the combination of mi-crowave and optical resonances.

The up-converted signal is processedphotonically by use of a tunable opticalfilter or local oscillator, and is then de-tected. Tunable optical filters can bemade to be frequency agile and to ex-hibit high resonance quality factors(high Q values), thereby making it possi-

ble to utilize a variety of signal-processingmodalities. Therefore, it is anticipatedthat when fully developed, receivers ofthis type will be compact and will be ca-pable of both wide-band and narrow-band signal processing. Thus, one com-pact receiver of this type would affordthe functionality that, heretofore, couldhave been obtained only by use of multi-ple heterodyne microwave receivers.

This work was done by Andrey Matsko,Lute Maleki, Vladimir Iltchenko, Nan Yu,Dmitry Strekalov, and Anatoliy Savchenkovof Caltech for NASA’s Jet Propulsion Labo-ratory. For more information, contact [email protected]. NPO-45313

A Microwave Signal Is Up-Converted to an optical signal, then filtered or otherwise processed pho-tonically before being detected.

WGMResonator/Modulator

Tunable Optical Filter orLocal Oscillator

ElectronicControlCircuitry

Microwave orLower-Frequency

Signal Out

Photodetector

Laser

Antenna

MicrowaveSignal In

ElectronicPreamplifier


Electronics/Computers

Interferometric synthetic apertureradar (InSAR) has been shown to pro-vide very sensitive measurements of sur-face deformation and displacement onthe order of 1 cm. Future systematicmeasurements of surface deformationwill require this capability over verylarge areas (300 km) from space. Toachieve these required accuracies, thesespaceborne sensors must exhibit lowtemporal decorrelation and be tempo-rally stable systems. An L-band (24-cm-wavelength) InSAR instrument using anelectronically steerable radar antenna issuited to meet these needs. In order toachieve the 1-cm displacement accuracy,the phased array antenna requiresphase-stable transmit/receive (T/R)modules. The T/R module operates atL-band (1.24 GHz) and has less than 1-deg absolute phase stability and lessthan 0.1-dB absolute amplitude stabilityover temperature. The T/R module isalso high power (30 W) and power effi-cient (60-percent overall efficiency).The design is currently implementedusing discrete components and surfacemount technology.

The basic T/R module architecture isaugmented with a calibration loop to

compensate for temperature variations,component variations, and path lossvariations as a function of beam settings.The calibration circuit consists of an am-plitude and phase detector, and othercontrol circuitry, to compare the meas-ured gain and phase to a reference sig-nal and uses this signal to control a pre-cision analog phase shifter and analogattenuator. An architecture was devel-oped to allow for the module to be bidi-rectional, to operate in both transmitand receive mode. The architecture alsoincludes a power detector used to main-tain a transmitter power output constantwithin 0.1 dB.

The use of a simple, stable, low-cost,and high-accuracy gain and phase de-

tector made by Analog Devices(AD8302), combined with a very-high-efficiency T/R module, is novel. Whilea self-calibrating T/R module capabilityhas been sought for years, a practicaland cost-effective solution has neverbeen demonstrated. By adding the cali-bration loop to an existing high-effi-ciency T/R module, there is a demon-strated order-of-magnitudeimprovement in the amplitude andphase stability.

This work was done by Constantine An-dricos and Wendy Edelstein of Caltech andVladimir Krimskiy of Santa Barbara Ap-plied Research for NASA’s Jet PropulsionLaboratory. Further information is containedin a TSP (see page 1).NPO-45147

L-Band Transmit/Receive Module for Phase-Stable Array AntennasA self-calibrating interferometric synthetic aperture radar instrument uses an electronicallysteerable radar antenna to achieve greater accuracy. NASA’s Jet Propulsion Laboratory, Pasadena, California

The photo shows the T/R Module on the front side and the 30-W Power Amp on the reverse side.

Microwave Power Combiner/Switch Utilizing a Faraday RotatorEither or both of two input ports could be coupled to one output port.NASA’s Jet Propulsion Laboratory, Pasadena, California

A proposed device for combining orswitching electromagnetic beams wouldhave three ports, would not contain anymoving parts, and would be switchableamong three operating states:• Two of the ports would be for input;

the remaining port would be for out-put.

• In one operating state, the signals atboth input ports would be coupledthrough to the output port.

• In each of the other two operatingstates, the signal at only one input port

would be coupled to the output port.The input port would be selectedthrough choice of the operating state.In one potential application, the device

would be used to switch or combine mi-crowave signals in a quasi-optical transmis-sion-line assembly that would be part of amillimeter-wave radar or telecommunica-tion system. In another potential applica-tion, a modified version of the devicewould be used to switch or combine lightsignals in a fiber-optic telecommunicationlink.

The two input ports would be config-ured to accommodate signals havingmutually orthogonal linear polariza-tions. A polarizer would be positioned tobisect the right angle formed by the lon-gitudinal axes of the input ports, and itspolarization would be oriented to so thatit would allow one input signal to passthrough and would reflect the otherinput signal. The orientations of theaforementioned components would besuch that after impinging on the polar-izer, both input signals would propagate


toward a three-state Faraday rotator. Thecomponents of the Faraday rotatorwould be a ferrite disk, a solenoidal elec-tromagnet coil for applying magneticbias, and two impedance-matchingplates — one on each side of the ferritedisk. The output port would be posi-tioned on the side opposite the inputside of the Faraday rotator and would beoriented to support polarization at anangle of 45° relative to both of the inputpolarizations.

The operating state would be selectedby adjusting the magnetic bias to selectone of three states of the Faraday rota-tor. In one state, the magnetic biaswould be set to cause the polarization ofa propagating signal to rotate throughan angle of +45° so as to allow one of the

input signals to propagate to the outputport. In another state, the direction ofthe magnetic bias would be set at the re-verse of that of the first-mentioned stateto obtain a polarization rotation of 45°,thereby allowing the other input signalto propagate to the output port.

The third state would be used for com-bining the powers of two mutually coher-ent input signals that, in an ideal case,would be of equal magnitude and woulddiffer in phase by 90°. In this state, themagnetic bias (and thus, the Faraday rota-tion) would be set to zero and the super-position of the input signals would resultin a 45°-polarized sum signal that wouldpropagate to the output port. In practice,because of magnetic hysteresis, this statecould not be obtained by simply abruptly

turning off the current in the electromag-net: It would be necessary to apply adamped sinusoidal excitation to the elec-tromagnet coil to effect degaussing.

This work was done by Raul Perez of Cal-tech for NASA’s Jet Propulsion Laboratory.


Innovative Technology Assets ManagementJPLMail Stop 202-2334800 Oak Grove DrivePasadena, CA 91109-8099E-mail: [email protected] to NPO-44316, volume and number


Compact Low-Loss Planar Magic-TThese wireless communications components are useful for base-station receivers, consumerelectronics, and industrial microwave instrumentation.Goddard Space Flight Center, Greenbelt, Maryland

This design allows broadband powercombining with high isolation betweenthe H port and E port, and achieves alower insertion loss than any otherbroadband planar magic-T. Passive mi-cro wave/millimeter-wave signal power iscombined both in-phase and out-of-phase at the ports, with the phase errorbeing less than ±1°, which is limited byport impedance.

The in-phase signal combiner consistsof two quarter-wavelength-long transmis-sion lines combined at the microstrip linejunction. The out-of-phase signal com-biner consists of two half-wavelength-longtransmission lines combined in series.Structural symmetry creates a virtual

ground plane at the combining junction,and the combined signal is convertedfrom microstrip line to slotline. Optimumrealizable characteristic impedances areused so that the magic-T provides broad-band response with low return loss.

The magic-T is used in microwave-and millimeter-wave frequencies, withthe operating bandwidth being approxi-mately 100 percent. The minimum isola-tion obtainable is 32 dB from port E toport H. The magic-T VSWR is less than1.1 in the operating band. Operatingtemperature is mainly dependent on thevariation in the dielectric constant of thesubstrate. Using crystallized substrate,the invention can operate in an ex-

tremely broad range of temperatures(from 0 to 400 K). It has a very high reli-ability because it has no moving partsand requires no maintenance, though itis desirable that the magic-T operate in alow-humidity environment. Fabricationof this design is very simple, using onlytwo metallized layers. No bond wires, viaholes, or air bridges are required. Addi-tionally, this magic-T can operate as anindividual component without auxiliarycomponents.

This work was done by Kongpop U-yen, Ed-ward J. Wollack, Terence Doiron, and SamuelH. Moseley of Goddard Space Flight Center.Further information is contained in a TSP (seepage 1). GSC-15353-1

Single-event upsets (SEUs) pose greatthreats to avionic systems’ state machinecontrol logic, which are frequently usedto control sequence of events and to qual-ify protocols. The risks of SEUs manifestin two ways: (a) the state machine’s stateinformation is changed, causing the state

machine to unexpectedly transition toanother state; (b) due to the asynchro-nous nature of SEU, the state machine’sstate registers become metastable, conse-quently causing any combinational logicassociated with the metastable registers tomalfunction temporarily. Effect (a) can

be mitigated with methods such as triple-modular redundancy (TMR). However,effect (b) cannot be eliminated and candegrade the effectiveness of any mitiga-tion method of effect (a).

Although there is no way to com-pletely eliminate the risk of SEU-in-

Using Pipelined XNOR Logic to Reduce SEU Risks in StateMachinesRisk is reduced by use of fast state-machine and error-detection logic.NASA’s Jet Propulsion Laboratory, Pasadena, California


The Next Generation Flight ControllerTrainer (NGFCT) is a relatively inexpen-sive system of hardware and software thatprovides high-fidelity training for space-shuttle flight controllers. NGFCT providessimulations into which are integrated thebehaviors of emulated space-shuttle vehi-cle onboard general-purpose computers(GPCs), mission-control center (MCC)displays, and space-shuttle systems as rep-resented by high-fidelity shuttle mission

simulator (SMS) mathematical models.The emulated GPC computers enable theexecution of onboard binary flight-specificsoftware. The SMS models include repre-sentations of system malfunctions that canbe easily invoked. The NGFCT softwarehas a flexible design that enables inde-pendent updating of its GPC, SMS, andMCC components.

This work was done by Scott Arnold,Matthew R. Barry, Isaac Benton, Michael M.

Bishop, Steven Evans, Jason Harvey, Timo-thy King, Jacob Martin, Al Mercier, WaltMiller, Dan L. Payne, Hanh Phu, James C.Thompson, and Ron Aadsen of United SpaceAlliance for Johnson Space Center. Further in-formation is contained in a TSP (see page 1).MSC-23617-1

Next Generation Flight Controller Trainer SystemLyndon B. Johnson Space Center, Houston, Texas

A quasi-optical transmission line(QOTL) has been developed as a low-losstransmission line for a spaceborne cloud-observing radar instrument that operatesat a nominal frequency of 94 GHz. ThisQOTL could also readily be redesigned foruse in terrestrial millimeter-wave radar sys-tems and millimeter-wave imaging systems.

In the absence of this or another low-loss transmission line, it would be nec-essary to use a waveguide transmissionline in the original radar application.Unfortunately, transmission losses in-crease and power-handling capacitiesof waveguides generally decrease withfrequency, such that at 94 GHz, the lim-itation on transmitting power and thecombined transmission and receptionlosses (> 5 dB) in a waveguide transmis-

sion line previously considered for theoriginal application would be unac-ceptable.

The QOTL functions as a very-low-loss, three-port circulator. The QOTLincludes a shaped input mirror that canbe rotated to accept 94-GHz transmitterpower from either of two high-poweramplifiers. Inside the QOTL, the trans-mitter power takes the form of a linearlypolarized beam radiated from a feedhorn. This beam propagates through asystem of mirrors, each of which refo-cuses the beam to minimize diffractionlosses. A magnetically biased ferrite discis placed at one of the foci to utilize theFaraday effect to rotate the polarizationof the beam by 45°. The beam is thentransmitted via an antenna system.

The radar return (scatter fromclouds, and/or reflections from otherobjects) is collected by the same an-tenna and propagates through the Fara-day rotator in the reverse of the direc-tion of propagation of the transmittedbeam. In the Faraday rotator, the polar-ization of the received signal is rotateda further 45°, so that upon emergingfrom the Faraday rotator, the receivedbeam is polarized at 90° with respect tothe transmitted beam. The transmittedand received signals are then separatedby a wire-grid polarizer.

This work was done by Raul M. Perez andWatt Veruttipong of Caltech for NASA’s JetPropulsion Laboratory. For more informa-tion, contact [email protected]. NPO-44236

Quasi-Optical Transmission Line for 94-GHz RadarThis apparatus functions as a very-low-loss, three-port circulator.NASA’s Jet Propulsion Laboratory, Pasadena, California

duced errors, the risk can be made verysmall by use of a combination of very faststate-machine logic and error-detectionlogic. Therefore, one goal of two mainelements of the present method is to de-sign the fastest state-machine logic cir-cuitry by basing it on the fastest genericstate-machine design, which is that of aone-hot state machine. The other of thetwo main design elements is to designfast error-detection logic circuitry and tooptimize it for implementation in a

field-programmable gate array (FPGA)architecture: In the resulting design, theone-hot state machine is fitted with amultiple-input XNOR gate for detectionof illegal states. The XNOR gate is im-plemented with lookup tables and withpipelines for high speed.

In this method, the task of designingall the logic must be performed manu-ally because no currently available logic-synthesis software tool can produce opti-mal solutions of design problems of this

type. However, some assistance is pro-vided by a script, written for this purposein the Python language (an object-ori-ented interpretive computer language)to automatically generate hardware de-scription language (HDL) code fromstate-transition rules.

This work was done by Martin Le, XinZheng, and Sunant Katanyoutant of Caltechfor NASA’s Jet Propulsion Laboratory. Furtherinformation is contained in a TSP (see page1). NPO-42401


Software

Converting From DDORSASF to APF

A computer program called“ddor_sasf2apf” converts delta-door(delta differential one-way range) re-quest from an SASF (spacecraft activitysequence file) format to an APF (apgenplan file) format for use in the Mars Re-connaissance Orbiter (MRO) mission-planning-and-sequencing process. TheAPF is used as an input to“APGEN/AUTOGEN” in the MRO activ-ity-planning and command-sequence-generating process to sequence thedelta-door (DDOR) activity. The DDORactivity is a spacecraft tracking techniquefor determining spacecraft location.

The input to ddor_sasf2apf is an inputrequest SASF provided by an observa-tion team that utilizes DDOR.ddor_sasf2apf parses this DDOR SASFinput, rearranging parameters and re-formatting the request to produce anAPF file for use in AUTOGEN and/orAPGEN. The benefit afforded byddor_sasf2apf is to enable the use of theDDOR SASF file earlier in the planningstage of the command-sequence-gener-ating process and to produce sequences,optimized for DDOR operations, thatare more accurate and more robust thanwould otherwise be possible.

This program was written by Roy E. Glad-den, Teerapat Khanampornpan, and ForestW. Fisher of Caltech for NASA’s Jet PropulsionLaboratory.

This software is available for commercial li-censing. Please contact Karina Edmonds ofthe California Institute of Technology at(626) 395-2322. Refer to NPO-45413.

Converting From CVF to AAFA computer program called “dsn config

converter” automates what had been amanual process for updating the multi-mission adaptation file (multi.aaf) usedby a multiple-mission-command-se -quence-generating process comprised ofa combination of the AUTOGEN andAPGEN programs mentioned in the im-mediately preceding article. The programconverts the dsn_config.cvf file that pro-vides DSN (Deep Space Network) an-tenna configuration code mappings froma context variable file (CVF) format usedin another part of the command genera-

tion process to an APGEN activity file(AAF) format used by AUTOGEN andAPGEN.

Whereas previously, the information inthe dsn_config.cvf file was manually en-coded into the multi.aaf file, now the pro-gram automatically generates adsn_config.aaf file from the dsn_config.cvffile. As part of this development effort themulti.aaf file was adapted to use the newdsn_config.aaf representations. Throughthis automation a tedious error-prone stephas now been replaced by a quick and ro-bust step.

This program was written by Roy E. Glad-den, Teerapat Khanampornpan, and ForestW. Fisher of Caltech for NASA’s Jet PropulsionLaboratory.


Documenting AUTOGENand APGEN Model Files

A computer program called “autogenhypertext map generator” satisfies a needfor documenting and assisting in visuali-zation of, and navigation through, modelfiles used in the AUTOGEN and APGENsoftware mentioned in the two immedi-ately preceding articles. This programparses autogen script files, autogenmodel files, PERL scripts, and apgen ac-tivity-definition files and produces a hy-pertext map of the files to aid in the nav-igation of the model. This program alsoprovides a facility for adding notes anddescriptions, beyond what is in the sourcemodel represented by the hypertext map.Further, this program provides access to asummary of the model through variable,function, sub routine, activity and re-source declarations as well as providingfull access to the source model andsource code. The use of the tool enableseasy access to the declarations and theability to traverse routines and calls whileanalyzing the model.

This program was written by Roy E. Glad-den, Teerapat Khanampornpan, Forest W.Fisher, and Chris C. Del Guercio of Caltechfor NASA’s Jet Propulsion Laboratory.


Sequence History Update Tool

The Sequence History Update Toolperforms Web-based sequence statisticsarchiving for Mars Reconnaissance Or-biter (MRO). Using a single UNIX com-mand, the software takes advantage ofsequencing conventions to automati-cally extract the needed statistics frommultiple files. This information is thenused to populate a PHP database, whichis then seamlessly formatted into a dy-namic Web page.

This tool replaces a previous tediousand error-prone process of manuallyediting HTML code to construct a Web-based table. Because the tool managesall of the statistics gathering and file de-livery to and from multiple data sourcesspread across multiple servers, there isalso a considerable time and effort sav-ings. With the use of The Sequence His-tory Update Tool what previously tookminutes is now done in less than 30 sec-onds, and now provides a more accu-rate archival record of the sequencecommanding for MRO.

This work was done by Teerapat Khanam-pornpan, Roy Gladden, Forest Fisher, andChris Del Guercio of Caltech for NASA’s JetPropulsion Laboratory.


Extraction and Analysis ofDisplay Data

The Display Audit Suite is an inte-grated package of software tools thatpartly automates the detection ofPortable Computer System (PCS) Dis-play errors. [PCS is a laptop computerused onboard the International SpaceStation (ISS).] The need for automa-tion stems from the large quantity ofPCS displays (6,000+, with 1,000,000+lines of command and telemetry data).The Display Audit Suite includes data-extraction tools, automatic error detec-tion tools, and database tools for gener-ating analysis spreadsheets.

These spreadsheets allow engineersto more easily identify many differentkinds of possible errors. The Suite sup-ports over 40 independent analyses,


and complements formal testing bybeing comprehensive (all displays canbe checked) and by revealing errorsthat are difficult to detect via test. Inaddition, the Suite can be run early inthe development cycle to find and cor-rect errors in advance of testing.

This software suite was developed by ChrisLand of The Boeing Company and KathrynMoye of the Dynacs Co. for Johnson SpaceCenter. Further information is contained in aTSP (see page 1). MSC-23630-1

MRO DKF Post-ProcessingTool

This software tool saves time and re-duces risk by automating two labor-in-tensive and error-prone post-processingsteps required for every DKF [DSN(Deep Space Network) Keyword File]that MRO (Mars Reconnaissance Or-biter) produces, and is being extendedto post-process the correspondingTSOE (Text Sequence Of Events) aswell. The need for this post-processingstep stems from limitations in the seq-gen modeling resulting in incorrectDKF generation that is then cleaned upin post-processing.

This work was done by Shanti Ayap ofLMCO, and Forest Fisher, Roy Gladden, andTeerapat Khanampornpan of NASA’s JetPropulsion Laboratory.


Rig Diagnostic ToolsRig Diagnostic Tools is a suite of ap-

plications designed to allow an opera-tor to monitor the status and health ofcomplex networked systems using aunique interface between Java applica-tions and UNIX scripts. The suite con-sists of Java applications, C scripts, Vx-Works applications, UNIX utilities, Cprograms, and configuration files.The UNIX scripts retrieve data fromthe system and write them to a certainset of files. The Java side monitorsthese files and presents the data inuser-friendly formats for operators touse in making troubleshooting deci-sions. This design allows for rapid pro-totyping and expansion of higher-leveldisplays without affecting the basicdata-gathering applications. The suiteis designed to be extensible, with theability to add new system components

in building block fashion without af-fecting existing system applications.This allows for monitoring of complexsystems for which unplanned shut-down time comes at a prohibitive cost.

This program was written by Kerry M.Soileau of Johnson Space Center and John W.Baicy of The Boeing Co. Further informationis contained in a TSP (see page 1). MSC-24158-1

MRO Sequence CheckingTool

The MRO Sequence Checking Toolprogram, mro_check, automates signifi-cant portions of the MRO (Mars Recon-naissance Orbiter) sequence checkingprocedure. Though MRO has similarchecks to the ODY’s (Mars Odyssey)Mega Check tool, the checks needed forMRO are unique to the MRO spacecraft.

The MRO sequence checking toolautomates the majority of the sequencevalidation procedure and check liststhat are used to validate the sequencesgenerated by MRO MPST (missionplanning and sequencing team). Thetool performs more than 50 differentchecks on the sequence. The automa-tion varies from summarizing dataabout the sequence needed for visualverification of the sequence, to per-forming automated checks on the se-quence and providing a report for eachstep. To allow for the addition of newchecks as needed, this tool is built in amodular fashion.

This work was done by Forest Fisher, RoyGladden, and Teerapat Khanampornpan ofCaltech for NASA’s Jet Propulsion Laboratory.


Science Activity Planner forthe MER Mission

The Maestro Science Activity Planneris a computer program that assists hu -man users in planning operations of theMars Explorer Rover (MER) missionand visualizing scientific data returnedfrom the MER rovers. Relative to itspredecessors, this program is more pow-erful and easier to use. This program isbuilt on the Java Eclipse open-sourceplatform around a Web-browser-baseduser-interface paradigm to provide anintuitive user interface to Mars roversand landers.

This program affords a combina-tion of advanced display and simula-tion capabilities. For example, a mapview of terrain can be generated fromimages acquired by the High Resolu-tion Imaging Science Explorer instru-ment aboard the Mars Recon -naissance Orbiter spacecraft andoverlaid with images from a naviga-tion camera (more precisely, a stereo-scopic pair of cameras) aboard arover, and an interactive, annotatedrover traverse path can be incorpo-rated into the overlay. It is also possi-ble to construct an overhead perspec-tive mosaic image of terrain fromnavigation-camera images.

This program can be adapted to simi-lar use on other outer-space missionsand is potentially adaptable to numer-ous terrestrial applications involvinganalysis of data, operations of robots,and planning of such operations for ac-quisition of scientific data.

This program was written by Jeffrey S. Nor-ris, Thomas M. Crockett, Jason M. Fox,Joseph C. Joswig, Mark W. Powell, KhawajaS. Shams, Recaredo J. Torres, Michael N.Wallick, and David S. Mittman of Caltechfor NASA’s Jet Propulsion Laboratory.


UAVSAR Flight-PlanningSystem

A system of software partly automatesplanning of a flight of the UninhabitedAerial Vehicle Synthetic ApertureRadar (UAVSAR) — a polarimetric syn-thetic-aperture radar system aboard anunpiloted or minimally piloted air-plane. The software constructs a flightplan that specifies not only the in-tended flight path but also the setup ofthe radar system at each point alongthe path.

A user first specifies the desiredimage swath by specifying certain geo-graphic and geometric features of theswath or the desired flight path. Usingan input digital elevation map (DEM),the software predicts the image swathand sets such variables as a data windowposition (DWP). A raster backscatterclassification file co-registered with theinput DEM can be used to estimateradar attenuation settings. The softwaredetermines whether such radar con-straints as those pertaining to duty cy-cles and data rates are obeyed, and de-


termines when radar settings should bemodified (for example, a DWPchanged, or gain changed in responseto a change in expected backscatter).The software constructs a Web page tofacilitate transfer of radar control files

and to provide access to KeyholeMarkup Language files, which can beused to display the flight path and asso-ciated information.

This program was written by Joanne G.Shimada, Anhua J. Chu, Elaine Chapin,

Scott Hensley, and Bruce D. Chapman of Cal-tech for NASA’s Jet Propulsion Laboratory.



Manufacturing & Prototyping

Templates for Deposition of Microscopic Pointed StructuresThese structures can be used as field emitters in plasma television screens.Goddard Space Flight Center, Greenbelt, Maryland

Templates for fabricating sharplypointed microscopic peaks arranged innearly regular planar arrays can be fabri-cated by a relatively inexpensive tech-nique that has recently been demon-strated. Depending on the intendedapplication, a semiconducting, insulat-ing, or metallic film could be depositedon such a template by sputtering, ther-mal evaporation, pulsed laser deposi-tion, or any other suitable conventionaldeposition technique. Pointed struc-tures fabricated by use of these tech-niques may prove useful as photocath-odes or field emitters in plasmatelevision screens. Selected peaks couldbe removed from such structures andused individually as scanning tips inatomic force microscopy or mechanicalsurface profiling.

The equipment and materials needed to form a template include the following:• Several small plates (e.g., microscope

slides) made of a suitable (preferablytransparent) rigid material such asglass, quartz, or sapphire;

• Two permanent magnets that pro-duce a flux density of the order of 1

kG (0.1 T) with an acceptably lowspatial variation over an area at leastas large as that of the template to beformed; and

• A ferrofluid (consisting of Fe2O3 par-ticles suspended in an oil-based solu-tion that includes a surfactant).A small quantity (≈1 mL) of the fer-

rofluid is either dropped or spun ontoone of the plates. The plate is orientedhorizontally, supported by two otherplates, positioned so that the ferrofluid-covered spot sits directly over one of themagnets. Next, using a combination ofthree other plates, the other magnet ispositioned a short distance above theferrofluid-covered spot (see Figure 1).The surface of the ferrofluid becomesdeformed into an array of peaks gener-ally oriented along magnetic-field lines.The positions of the magnets relative toeach other and to the ferrofluid are ad-justed to minimize nonuniformity in thearray of peaks and to maximize the as-pect ratio of the peaks. Next, the fer-rofluid is dried in room air using a gen-tly blowing muffin fan. The ferrofluid isthen further dried and hardened in athermal evaporation chamber pumped

down to a pressure of about 10–6 torr(about 1.3 × 10–4 Pa).

The resulting structure is ready foruse as a template for deposition. For ex-ample, Figure 2 shows selected views ofsuch a template that has been coatedwith thermally evaporated silver.

This work was done by Diane E. Pugel ofGoddard Space Flight Center. Further infor-mation is contained in a TSP (see page 1).

This invention is owned by NASA, and apatent application has been filed. Inquiriesconcerning nonexclusive or exclusive licensefor its commercial development should be ad-dressed to the Patent Counsel, Goddard SpaceFlight Center, (301) 286-7351. Refer to GSC-14871-1.

Figure 1. A Spot of Ferrofluid on a Plate is positioned between two magnets. The magnetic field de-forms the upper surface of the ferrofluid into an array of peaks.

Magnet

Magnet

Ferrofluid

Glass

Figure 2. A Portion of a Template fabricated asdescribed in the text is shown here in top-viewand side-view scanning electron micrographs.

TOP VIEW

SIDE VIEW


Materials

Adjustable Membrane Mirrors Incorporating G-ElastomersNASA’s Jet Propulsion Laboratory, Pasadena, California

Lightweight, flexible, large-aperturemirrors of a type being developed foruse in outer space have unimorphstructures that enable precise adjust-ment of their surface figures. A mirrorof this type includes a reflective mem-brane layer bonded with an electrostric-tive grafted elastomer (G-elastomer)layer, plus electrodes suitably posi-tioned with respect to these layers. Byvirtue of the electrostrictive effect, anelectric field applied to the G-elastomermembrane induces a strain along themembrane and thus causes a deflection

of the mirror surface. Utilizing this ef-fect, the mirror surface figure can beadjusted locally by individually address-ing pairs of electrodes.

G-elastomers, which were developed atNASA Langley Research Center, werechosen for this development in prefer-ence to other electroactive polymerspartly because they offer superior electro-mechanical performance. Whereas otherelectroactive polymers offer, variously,large strains with low moduli of elasticityor small strains with high moduli of elas-ticity, G-elastomers offer both large strains

(as large as 4 percent) and high moduli ofelasticity (about 580 MPa). In addition, G-elastomer layers can be made by standardmelt pressing or room-temperature solu-tion casting.

This work was done by Zensheu Changand Rhonda M. Morgan of Caltech, Eui-Hyeok Yang of Stevens Institute of Technol-ogy, Yoshikazu Hishinuma of Fuji FilmCorp., and Ji Su and Tian-Bing Xu ofNASA Langley Research Center for NASA’sJet Propulsion Laboratory. For more informa-tion, contact [email protected]. NPO-45616

Hall-Effect Thruster Utilizing Bismuth as PropellantMarshall Space Flight Center, Alabama

A laboratory-model Hall-effect space-craft thruster was developed that uti-lizes bismuth as the propellant. Xenonwas used in most prior Hall-effectthrusters. Bismuth is an attractive alter-native because it has a larger atomicmass, a larger electron-impact-ioniza-tion cross-section, and is cheaper andmore plentiful.

The design of this thruster includesmultiple temperature-control zones andother features that reduce parasiticpower losses. Liquid bismuth (which

melts at a temperature of 271°C) is sup-plied by a temperature-controlled reser-voir to a vaporizer. The vaporizer ex-hausts to an anode/gas distributorinside a discharge channel that consistsof a metal chamber upstream of ceramicexit rings. In the channel, bismuth ionsare produced through an electron im-pact ionization process and acceleratedas in other Hall-effect thrusters. The dis-charge region is heated by the dischargeand an auxiliary anode heater, which isrequired to prevent bismuth condensa-

tion at low power levels and at thrusterstart-up. A xenon discharge is also usedfor preheating the discharge channel,but an anode heater could provideenough power to start the bismuth dis-charge directly.

This work was done by James Szabo,Charles Gasdaska, Vlad Hruby, and MikeRobin of Busek Co., Inc. for Marshall SpaceFlight Center. For further information, con-tact Sammy Nabors, MSFC Commerciali -zation Assistance Lead, at [email protected]. Refer to MFS-32440-1.

High-Temperature Crystal-Growth Cartridge Tubes Made by VPSMechanical properties and maximum useful temperature exceed those of tungsten-alloy tubes. Marshall Space Flight Center, Alabama

Cartridge tubes for use in a crystal-growth furnace at temperatures as highas 1,600°C have been fabricated by vac-uum plasma spraying (VPS). These car-tridges consist mainly of an alloy of 60weight percent molybdenum with 40weight percent rhenium, made frommolybdenum powder coated with rhe-nium. This alloy was selected because ofits high melting temperature

(≈2,550°C) and because of its excellentductility at room temperature. Thesecartridges are intended to supplanttungsten/nickel-alloy cartridges, whichcannot be used at temperatures above≈1,300°C.

Graphite mandrels were used assubstrates for VPS to form the car-tridge tubes to the desired size andshape. A mandrel was placed in the

VPS chamber, oriented vertically. Be-fore spraying, the plasma gun wasused to heat the mandrel to a temper-ature of about 1,093°C. Then, theMo/Re alloy precursor powder wasdeposited by VPS on the mandrel to athickness between 0.51 and 0.64 mm.The deposition was done in one pass,spraying from the top to the bottomof the mandrel.


Then a tantalum coat was depositedin a similar manner onto the Mo/Redeposit to a thickness between 0.13 and0.18 mm. The tantalum coat serves as asealing layer, increasing the protectionof the Mo/Re alloy against the forma-tion of such volatile oxides as MoO3.

Next, the pressure in the chamberwas reduced to <100 mtorr (less thanabout 13 Pa) and the cartridge allowedto cool. Once the cartridge had cooledto room temperature, the chamber was

opened to the atmosphere and the car-tridge was removed from the mandrel.

A cross section of a representativecartridge tube fabricated in thisprocess showed a good bond betweenthe tantalum coat and the main bodyof Mo/Re alloy. Both the Mo/Re andthe Ta were dense. Because this tubewas not heat treated, the Mo/Re-alloylayer still contained two phases — oneMo-rich and one Re-rich. Tests of themechanical properties of tubes like this

one in the as-sprayed condition haverevealed a vast improvement over simi-lar tungsten-alloy tubes in the as-sprayed condition.

This work was done by Richard Holmes of Marshall Space Flight Center and Scott O’Dell, Timothy McKechnie, andChristopher Power of Plasma Processes,Inc. For further information, contactSammy Nabors, MSFC CommercializationAssistance Lead, at [email protected]. MFS-31540-1

Improved crucibles consisting mainly ofmetal-reinforced ceramic ampules havebeen developed for use in experiments inwhich material specimens are heated inthe crucibles to various high tempera-tures, then quenched by, for example,plunging the crucibles into water at roomtemperature. A quench crucible of the tra-ditional type intended to be supplanted bythe improved crucibles consists mainly of aceramic or graphite ampule inside a metalcartridge, with a gap between the metaland the cartridge, as shown on the left sideof the figure.

The need for the improved quenchcrucibles arises as follows: In a tradi-tional quench crucible, the gap be-tween the ampule and the metal car-tridge impedes the transfer of heat tosuch a degree that the quench rate (therate of cooling of the specimen) can betoo low to produce the desired effect inthe specimen. One can increase thequench rate by eliminating the metalcartridge to enable direct quenching ofthe ampule, but then the thermal shockof direct quenching causes cracking ofthe ampule.

In a quench crucible of the presentimproved type, there is no gap andno metal cartridge in the traditionalsense. Instead, there is an overlay ofmetal in direct contact with the am-pule, as shown on the right side ofthe figure. Because there is no gapbetween the metal overlay and theampule, the heat-transfer rate can bemuch greater than it is in a tradi-tional quench crucible. The metaloverlay also reinforces the ampuleagainst cracking.

The choice of ampule material andmetal depends on the specific applica-

tion. In general, the ampule materialshould be chemically compatible with thespecimen material. The overlay metalshould be chosen to have a coefficient ofthermal expansion (CTE) as close as pos-sible to that of the ampule material. Ex-amples of suitable ampule/metal-overlaymaterial pairs include the following:• graphite (CTE = 8.0 × 10–6 K–1) and

stainless steel (CTE = 9.9 × 10–6 K–1)• aluminum nitride (CTE = 5.2 × 10–6

K–1) and tungsten heavy alloy (CTE =5.0 × 10–6 K–1) and

• silicon carbide (CTE = 4.5 × 10–6 K–1)

and tungsten heavy alloy (CTE = 5.0 ×10–6 K–1).Several thermal-spray pro cesses for

applying metal overlays to ampules wereinvestigated. Of these processes, vacuumplasma spraying was found to yield thebest results.

This work was done by Richard R. Holmesand Edgar Carrasquillo of Marshall SpaceFlight Center and J. Scott O’Dell and TimothyN. McKechnie of Plasma Processes Inc. Forfurther information, contact Sammy Nabors,MSFC Commercialization Lead, [email protected]. MFS-31598-1

Quench Crucibles Reinforced With MetalSpecimens can be quenched rapidly, without cracking ampules.Marshall Space Flight Center, Alabama

AdjustmentBlock

Ampoule

Gap

MetalCartridge

AdjustmentBlock

MetalOverlay

TRADITIONAL QUENCH CRUCIBLE IMPROVED QUENCH CRUCIBLE

The Metal Cartridge and Gap surrounding the ampule are replaced with an overlay of metal in inti-mate contact with the ampule.


Mechanics/Machinery

Deep-Sea Hydrothermal-Vent SamplerThis apparatus collects hydrothermal-plume samples uncontaminated by surrounding water.NASA’s Jet Propulsion Laboratory, Pasadena, California

An apparatus is being developed forsampling water for signs of microbial lifein an ocean hydrothermal vent at adepth of as much as 6.5 km. Heretofore,evidence of microbial life in deep-sea hy-drothermal vents has been elusive anddifficult to validate. Because of the ex-treme conditions in these environments(high pressures and temperatures oftenin excess of 300°C), deep-sea hydrother-mal-vent samplers must be robust. Be-cause of the presumed low density ofbiomass of these environments, sam-plers must be capable of collecting watersamples of significant volume. It is alsoessential to prevent contamination ofsamples by microbes entrained from sur-rounding waters. Prior to the develop-ment of the present apparatus, no sam-pling device was capable of satisfyingthese requirements.

The apparatus (see figure) includes anintake equipped with a temperatureprobe, plus several other temperatureprobes located away from the intake. Thereadings from the temperature probesare utilized in conjunction with readingsfrom flowmeters to determine the posi-tion of the intake relative to the hy-drothermal plume and, thereby, to posi-tion the intake to sample directly from theplume. Because it is necessary to collectlarge samples of water in order to obtainsufficient microbial biomass but it is notpractical to retain all the water from thesamples, four filter arrays are used to con-centrate the microbial biomass (which isassumed to consist of particles larger than0.2 µm) into smaller volumes. The appa-

ratus can collect multiple samples perdive and is designed to process a total vol-ume of 10 L of vent fluid, of which mostpasses through the filters, leaving a totalpossibly-microbe-containing sample vol-ume of 200 mL remaining in filters.

A rigid titanium nose at the intake isused for cooling the sample water beforeit enters a flexible inlet hose connected toa pump. As the water passes through thetitanium nose, it must be cooled to a tem-perature that is above a mineral-precipita-tion temperature of 100°C but below theupper working temperature (230°C) ofswitching valves and tubes in the appara-tus. The sample water then passes into amanifold tube, from whence the switch-ing valves can direct the water through ei-ther a bypass tube or any one of the filterarrays, without contamination from a pre-vious sample. Each filter array consists ofseries of filters having pore sizes decreas-ing in the direction of flow: 90-, 60-, 15-,and 7-µm prefilters and a large-surface-area 0.2-µm collection filter. All the filtertaps are located between the intake andthe bypass tube so that each time the by-pass tube is used, the entire manifold tubeis flushed as well.

The switching valves include five passiveones (a check valve for the bypass tube foreach of four filter arrays) at the upstream(manifold) end and an active one (a five-position actuated valve) at the down-stream end. The incorporation of thecheck valves at the upstream end makes itunnecessary to use actuated valves at bothends. Once the actuated valve has beenturned to the bypass position, the pump

begins to flush the intake and manifold ofany particulate matter that may have accu-mulated. After flushing, sampling isstarted by setting the actuated valve to passwater through one of the filter arrays. Theprocess of flushing and sampling is re-peated for each of the four filter arrays.

Because the apparatus is rated to adepth of 6.5 km, it is pressure-compen-sated; as the pressure increases betweenatmospheric at the ocean surface andabout 10 kpsi (≈69 MPa) at maximumdepth, the volume of water within thesystem decreases by about 2.7 percent. Aflexible membrane within the apparatusaccommodates the compression and ex-pansion of water upon descent and as-cent, respectively.

The apparatus includes a system formonitoring and regulating tempera-tures, pressures, and flow rates through-out the system. This control and moni-toring system includes a smallmicroprocessor, motor controllers (forthe pump and the valve actuator), theaforementioned temperature probes,pressure sensors, and a serial data link toa laptop computer aboard a submarineor other vessel used to bring the appara-tus to and from the hydrothermal vent.

This work was done by Alberto E. Behar,Kasthur Venkateswaran, and Jaret B.Matthews o f Cal t ech and Cesar Ri-vadeneyra, James C. Bruckner, Edmond So,and Goran Basic of the International SpaceUniversity for NASA’s Jet Propulsion Labo-ratory. Further information is contained in aTSP (see page 1).NPO-42617

Manifold Tubes andCheck Valves

Bypass Tube Filter Arrays

Filter-InletTubes

Rigid Cooling Nose

Intake

Flexible Inlet Tube

Temperature Probe

Temperature Probe

Water Out(Dumped to Sea)

ValveActuator

Pump

The Intake Is Positioned in the plume of a hydrothermal vent with the help of temperature-probe readings, then water is pumped from the intake throughfilter arrays to collect microbes.


Mars Rocket Propulsion SystemMarshall Space Flight Center, Alabama

A report discusses the methane andcarbon monoxide/LOX (McLOx)rocket for ascent from Mars as well asother critical space propulsion tasks.The system offers a specific impulse over370 s —– roughly 50 s higher than exist-ing space-storable bio-propellants. Cur-rent Mars in-situ propellant production(ISPP) technologies produce impuremethane and carbon monoxide in vari-ous combinations. While separation andpurification of methane fuel is possible,it adds complexity to the propellant pro-duction process and discards an other-wise useful fuel product. The McLOxmakes such complex and wasteful

processes unnecessary by burning themethane/CO mixtures produced by theMars ISPP systems without the need forfurther refinement.

Despite the decrease in rocket-spe-cific impulse caused by the CO admix-ture, the improvement offered by con-comitant increased propellant densitycan provide a net improvement in stageperformance. One advantage is the in-crease of the total amount of propellantproduced, but with a decrease in massand complexity of the required ISPPplant. Methane/CO fuel mixtures alsomay be produced by reprocessing theorganic wastes of a Moon base or a

space station, making McLOx engineskey for a human Lunar initiative or theInternational Space Station (ISS) pro-gram. Because McLOx propellant com-ponents store at a common tempera-ture, very lightweight and compactcommon bulkhead tanks can be em-ployed, improving overall stage per-formance further.

This work was done by Robert Zubrinand Dan Harber of Pioneer Astronautics for Marshall Space Flight Center. For further information, contact SammyNabors, MSFC Commercialization Lead, [email protected]. MFS-32541-1

Two-Stage Passive Vibration IsolatorNASA’s Jet Propulsion Laboratory, Pasadena, California

The design and testing of a structuralsystem were implemented to hold theoptics of the planned Space Interferom-etry Mission (SIM) at positions and ori-entations characterized by vibrationaltranslation and rotation errors of nomore than a few nanometers or a fewmilliarcseconds, respectively. Much ofthe effort was devoted to a test bed forverifying the predicted behavior of a vi-bration-isolation structural subsystemworking together with an active controlsystem for positioning and orienting theSIM optics.

There was considerable emphasis on the vibration-isolation subsystem,which was passive and comprised two stages. The main sources of vibration were six reaction wheels in anassembly denoted the “backpack.” Thefirst vibration-isolation stage consistedof hexapod isolator mounts — one foreach reaction wheel — characterized bya natural vibration frequency of 10 Hz.The second stage was a set of threebeams, disposed between the backpackand the structure that held the SIM op-tics, that were flexured such that they

transmitted only bending loads, with anatural vibrational frequency anddamping of about 5 Hz and 4 percent,respectively. Preliminary test resultswere presented and characterized asdemonstrating the effectiveness of thetwo-stage vibration-isolation design.

This work was done by Renaud Goul-lioud, Yekta Gursel, Timothy Neville, AllenJ. Bronowicki, David Platus, and RhondaMacDonald of Caltech for NASA’s JetPropulsion Laboratory. For more informa-tion, contact [email protected]

A compression tool used to make 1-in.(2.5-cm) diameter disks of high-temper-ature polymers was designed to beshorter and of larger diameter than con-ventional tools to reduce heat loss to thesurrounding air, thus making more effi-cient use of applied heat. This system isless sensitive to the amount and qualityof insulation than previous tools, pro-vides more repeatable processing, andimproves the quality of the samples pro-duced. These improvements come with-out increasing the weight of the punchportion of the tool over that of a conven-tional version, an important qualitywhen handling lower-viscosity resins.

In use, the base and body of thetool are assembled, and the polymerto be processed is placed into thebody of the tool. The punch is in-serted, and the assembled tool isplaced into the press. A tempera-ture/pressure profile appropriate tothe nature of the polymer is applied.A series of computational and exper-imental runs were made using a con-ventional tool to validate the compu-tational model. The new tool designwas then modeled, and when the per-formance showed a marked improve-ment, the new tool was manufac-tured. A new series of experimental

runs showed that the thermal per-formance of the new tool agreed wellwith model predictions.

This work was done by Maria A.Kuczmarski and James C. Johnston ofGlenn Research Center and DeNiseHardy-Green of the University of Akron.Further information is contained in aTSP (see page 1).

Inquiries concerning rights for the com-mercial use of this invention should be ad-dressed to NASA Glenn Research Center, In-novative Partnerships Office, Attn: SteveFedor, Mail Stop 4–8, 21000 BrookparkRoad, Cleveland, Ohio 44135. Refer toLEW-17990-1.

Improved Thermal Design of a Compression MoldJohn H. Glenn Research Center, Cleveland, Ohio


An enhanced version of the scheme re-ported in “Pseudo-Waypoint Guidance forProximity Spacecraft Maneuvers” (NPO-42753), NASA Tech Briefs, Vol. 31, No. 6(June 2007), page 73 was developed. Torecapitulate: the scheme provides algo-rithms for guidance and control (G&C)of a spacecraft maneuvering near a smallastronomical body. The open-loop guid-ance problem is solved in advance or inreal time by use of the pseudo-waypointgeneration (PWG) method. Feedbackcontrol is implemented to track PWG tra-

jectories, in a manner that enables updat-ing of G&C in a model-predictive manner.The scheme includes silent periods fol-lowing each thruster firing.

The original version of the scheme pro-vides for a fire-first, followed-by-silence se-quence, which is disadvantageous in thatthe silence after final firing precludes re-duction of any remaining velocity error —an unacceptable result in the case of a ma-neuver for which a specific final velocity isrequired. In the enhanced version, thescheme is augmented with a fire-second

technique, so that the final velocity can beestablished with a much higher precisionbecause both the guidance and feedbackfiring can be performed and ceased at thefinal maneuver time.

This work was done by John Carson and Be-hçet Açikmese of Caltech for NASA’s Jet Propul-sion Laboratory.

The software used in this innovation isavailable for commercial licensing. Please con-tact Karina Edmonds of the California Insti-tute of Technology at (626) 395-2322. Refer toNPO-44276.

Enhanced Pseudo-Waypoint Guidance for Spacecraft ManeuversNASA’s Jet Propulsion Laboratory, Pasadena, California


Physical Sciences

A Global Positioning System (GPS)-reflection/occultation interferometrywas examined as a means of altimetry ofwater and ice surfaces in polar regions.In GPS-reflection/occultation interfer-ometry, a GPS receiver aboard a satellitein a low orbit around the Earth is usedto determine the temporally varying car-rier-phase delay between (1) one com-ponent of a signal from a GPS transmit-ter propagating directly through theatmosphere just as the GPS transmitterfalls below the horizon and (2) another

component of the same signal, propa-gating along a slightly different path, re-flected at glancing incidence upon thewater or ice surface.

The integer-cycle phase-differenceambiguity is resolved by noting thatboth signal components eventually col-lapse into a single component, repre-senting zero phase difference. Fromthe phase difference and the known po-sitions of the two spacecraft as func-tions of time, an atmospheric correc-tion obtained as the main data product

of the GPS-receiver mission, and basicgeometry, the difference in length be-tween the direct and reflection signalpaths and the altitude of the effectivespecular-reflection point can be calcu-lated. This method yields altitude atabout 0.7-m precision with horizontalresolution of a few kilometers.

This work was done by Estel Cardellach,Manuel De La Torre, George A. Hajj, and ChiAo of Caltech for NASA’s Jet Propulsion Labo-ratory. For more information, contact [email protected]. NPO-41551

Altimetry Using GPS-Reflection/Occultation InterferometryNASA’s Jet Propulsion Laboratory, Pasadena, California

Thermally Driven Josephson EffectNASA’s Jet Propulsion Laboratory, Pasadena, California

A concept is proposed of the ther-mally driven Josephson effect in super-fluid helium. Heretofore, the Josephsoneffect in a superfluid has been recog-nized as an oscillatory flow that arises inresponse to a steady pressure differencebetween two superfluid reservoirs sepa-rated by an array of submicron-sized ori-fices, which act in unison as a singleJosephson junction. Analogously, thethermally driven Josephson effect is anoscillatory flow that arises in response toa steady temperature difference.

The thermally driven Josephson ef-fect is partly a consequence of a quan-tum-mechanical effect known as thefountain effect, in which a tempera-ture difference in a superfluid is ac-companied by a pressure difference.The thermally driven Josephson effectmay have significance for the develop-ment of a high-resolution gyroscopebased on the Josephson effect in a su-perfluid: If the pressure-driven Joseph-son effect were used, then the fluid onthe high-pressure side would become

depleted, necessitating periodic inter-ruption of operation to reverse thepressure difference. If the thermallydriven Josephson effect were used,there would be no net flow and so theoscillatory flow could be maintainedindefinitely by maintaining the re-quired slightly different temperatureson both sides of the junction.

This work was done by Konstantin Penanenand Talso Chui of Caltech for NASA’s JetPropulsion Laboratory. For more information,contact [email protected]. NPO-40231

A computational-simulation study hasbeen presented of effects of perturba-tion wavelengths and initial Reynoldsnumbers on the transition to turbulenceof a heptane/nitrogen mixing layer atsupercritical pressure. The governingequations for the simulations were thesame as those of related prior studies re-ported in NASA Tech Briefs. Two-dimen-sional (2D) simulations were performedwith initially im posed span wise perturba-tions whereas three-dimensional (3D)simulations had both streamwise andspanwise initial perturbations.

The 2D simulations were undertakento ascertain whether perturbations hav-ing the shortest unstable wavelengthobtained from a linear stability analysisfor inviscid flow are unstable in viscousnonlinear flows. The goal of the 3Dsimulations was to ascertain whetherperturbing the mixing layer at differ-ent wavelengths affects the transitionto turbulence.

It was found that transitions to turbu-lence can be obtained at different per-turbation wavelengths, provided thatthey are longer than the shortest unsta-

ble wavelength as determined by 2D lin-ear stability analysis for the inviscid caseand that the initial Reynolds number isproportionally increased as the wave-length is decreased. The transitionalstates thus obtained display different dy-namic and mixture characteristics, de-parting strongly from the be haviors ofperfect gases and ideal mixtures.

This work was done by Nora Okong’o andJosette Bellan of Caltech for NASA’s Jet Propul-sion Laboratory. For more information, [email protected]

Perturbation Effects on a Supercritical C7H16/N2 Mixing LayerNASA’s Jet Propulsion Laboratory, Pasadena, California


Gold Nanoparticle Labels Amplify Ellipsometric SignalsMarshall Space Flight Center, Alabama

The ellipsometric method reported inthe immediately preceding article wasdeveloped in conjunction with a methodof using gold nanoparticles as labels onbiomolecules that one seeks to detect.The purpose of the labeling is to exploitthe optical properties of the goldnanoparticles in order to amplify themeasurable ellipsometric effects andthereby to enable ultrasensitive detec-tion of the labeled biomolecules withoutneed to develop more-complex ellipso-metric instrumentation.

The colorimetric, polarization, light-scattering, and other optical properties ofnanoparticles depend on their sizes andshapes. In the present method, these size-

and-shape-dependent properties are usedto magnify the polarization of scatteredlight and the diattenuation and retar-dance of signals derived from ellipsome-try. The size-and-shape-dependent opticalproperties of the nanoparticles make itpossible to interrogate the nanoparticlesby use of light of various wavelengths, asappropriate, to optimally detect particlesof a specific type at high sensitivity.

Hence, by incorporating gold nano -particles bound to biomolecules as pri-mary or secondary labels, the perform-ance of ellipsometry as a means ofdetecting the biomolecules can be im-proved. The use of gold nanoparticles aslabels in ellipsometry has been found to

afford sensitivity that equals or exceedsthe sensitivity achieved by use of fluores-cence-based methods. Potential applica-tions for ellipsometric detection of gold-nanoparticle-labeled biomoleculesinclude monitoring molecules of inter-est in biological samples, in-vitro diag-nostics, process monitoring, general en-vironmental monitoring, and detectionof biohazards.

This work was done by Sr ivatsaVenkatasubbarao of Intelligent Optical Systems,Inc. for Marshall Space Flight Center. For fur-ther information, contact Sammy Nabors,MSFC Commercialization Assistance Lead, [email protected]. Refer to MFS-32507-1.

Phase Matching of Diverse Modes in a WGM ResonatorPhase matching is necessary for exploitation of nonlinear optical phenomena.NASA’s Jet Propulsion Laboratory, Pasadena, California

Phase matching of diverse electro-magnetic modes (specifically, coexist-ing optical and microwave modes) in awhispering-gallery-mode (WGM) res-onator has been predicted theoreti-cally and verified experimentally. Suchphase matching is necessary for storageof microwave/terahertz and opticalelectromagnetic energy in the sameresonator, as needed for exploitationof nonlinear optical phenomena.

WGM resonators are used in re -search on nonlinear optical phenom-ena at low optical intensities and as a basis for design and fabrication of novel optical devices. Examples ofnonlinear optical phenomena re -cently demonstrated in WGM res-onators include low-threshold Ramanlasing, optomechanical oscillations,frequency doubling, and hyperpara-metric oscillations.

The present findings regardingphase matching were made in researchon low-threshold, strongly nondegen-erate parametric oscillations in lithiumniobate WGM resonators. The princi-ple of operation of such an oscillator isrooted in two previously observed phe-nomena: (1) stimulated Raman scatter-ing by polaritons in lithium niobateand (2) phase matching of nonlinearoptical processes via geometrical con-finement of light. The oscillator is

partly similar to terahertz oscillatorsbased on lithium niobate crystals, thekey difference being that a novel geo-metrical configuration of this oscillatorsupports oscillation in the continuous-wave regime. The high resonance qual-ity factors (Q values) typical of WGM

resonators make it possible to achieveoscillation at a threshold signal levelmuch lower than that in a non-WGM-resonator lithium niobate crystal.

The applicable theory states that theparametric interaction takes place in aWGM resonator if the photon-energy-

Nonlinear Optical Phenomena are excited in a WGM resonator disk and the output spectrum is meas-ured to obtain evidence of those phenomena. In the phenomenon of particular interest here, an op-tical pump photon of wave vector kp is scattered into an optical signal photon of wave vector ks anda microwave idler photon of wave vector ki. The idler photon is not necessarily confined within WGMresonator if its wavelength exceeds the thickness of the resonator disk.

ENLARGED SLANT VIEW OF WGM RESONATOR

SCHEMATIC DIAGRAM OF APPARATUS

LaserPolarizationController

WGMResonator

OpticalFibers

1:9 Fiber-OpticPower Splitter

GRIN Lenses

Oscilloscope

Photodiode

OpticalSpectrumAnalyzer

ks

kp


conservation law and the phase-match-ing condition are satisfied. The pho-ton-energy-conservation law can bestated simply as ωp = ωs + ωi, where ω isproportional to the frequency or en-ergy of the photon denoted by its sub-script and p, s, and i denote the pump,signal, and idler frequencies, respec-tively. The phase-matching condition issatisfied if the volume integral of theproduct of the complex amplitudes ofthe pump, signal, and idler electromag-netic fields differs from zero.

In the general case, phase matchingof an optical field with a microwavefield cannot be achieved in a WGM res-onator because the indices of refrac-tion of the bulk resonator material aredifferent in the optical and microwavefrequency ranges. However, the theory

also shows that it is possible to tailorthe spatial structures of the WGMmodes, so as to obtain phase matchingof fields at resonance frequencies thatsatisfy the photon-energy-conservationlaw, through appropriate tailoring ofthe size and shape of the WGM res-onator. This is equivalent to matchingof effective indices of refraction for thepump, signal, and idler fields.

Evidence that phase matching can beachieved through suitable choice ofsize and shape was obtained in experi-ments on an apparatus depictedschematically in the figure. In each ex-periment, laser light centered at awavelength of ≈1,319 nm or ≈1,559 nmwas sent through a polarization con-troller, a grating-index-of-refraction(GRIN) lens, and a diamond prism

into a lithium niobate WGM resonator,and light was coupled out of the WGMresonator through the diamond prism,another GRIN lens, and optical fibersto a photodiode and an optical spectralanalyzer. In one experiment, the spec-trum of light coming out of the WGMresonator was found to include side-bands associated with strongly nonde-generate parametric oscillations thathad been predicted theoretically. Inother experiments, oscillations with,variously, confined or unconfined idlerfields were observed.

This work was done by AnatoliySavchenkov, Dmitry Strekalov, Nan Yu, An-drey Matsko, Makan Mohageg, and LuteMaleki of Caltech for NASA’s Jet PropulsionLaboratory. Further information is containedin a TSP (see page 1). NPO-45120

WGM Resonators for Terahertz-to-Optical Frequency ConversionReceivers containing these devices are contemplated for astronomical and military uses.NASA’s Jet Propulsion Laboratory, Pasadena, California

Progress has been made toward solvingsome practical problems in the imple-mentation of terahertz-to-optical fre-quency converters utilizing whispering-gallery-mode (WGM) resonators. Suchfrequency converters are expected to beessential parts of non-cryogenic tera-hertz-radiation receivers that are, vari-ously, under development or contem-plated for a variety of applications inairborne and spaceborne instrumenta-tion for astronomical and military uses.

In most respects, the basic principlesof terahertz-to-optical frequency con-version in WGM resonators are thesame as those of microwave (sub-tera-hertz)-to-optical frequency conversionin WGM resonators, various aspects ofwhich were discussed in the three pre-ceeding articles. To recapitulate: In areceiver following this approach, a pre-amplified incoming microwave signal(in the present case, a terahertz signal)is up-converted to an optical signal by atechnique that exploits the nonlinear-ity of the electromagnetic response of awhispering-gallery-mode (WGM) res-onator made of LiNbO3 or anothersuitable electro-optical material. Up-conversion takes place by three-wavemixing in the resonator. To ensure therequired interaction among the opticaland terahertz signals, the WGM res-onator must be designed and fabri-cated to function as an electro-optical

modulator while simultaneously ex-hibiting (1) resonance at the re-quired microwave and optical operat-ing frequencies and (2) phasematching among the microwave andoptical signals circulating in the res-onator. Downstream of the WGM res-

onator, the up-converted signal isprocessed photonically by use of a tun-able optical filter or local oscillator andis then detected.

The practical problems addressed inthe present development effort are thefollowing:

WGM Resonator Ring

Low-Index-of-Refraction Post

A WGM Resonator Ring is mounted on a post made of a material having an index of refraction sig-nificantly lower than that of the ring to provide mechanical support without sacrificing confinementof the WGM modes in the ring.


• Satisfaction of the optical and terahertzresonance-frequency requirement is astraightforward matter, inasmuch as theoptical and terahertz spectra can bemeasured. However, satisfaction of thephase-matching requirement is moredifficult. The approach followed in thepresent development is to performcomputer simulations of the microwaveand optical signals circulating in the res-onator to test for phase matching.

• To enable excitation of the terahertzWGM resonator mode, it is also nec-essary to ensure phase matching be-tween that mode and the incomingterahertz radiation. In the present

development, the incoming signal iscoupled into the WGM resonator viaa tapered waveguide in the form of afused silica rod. The phase-matchingrequirement is satisfied at one pointalong the taper; the rod is positionedwith this point in proximity to theWGM resonator.

• To maximize the conversion effi-ciency, it is necessary to maximize thespatial overlap among the terahertzand optical modes in the WGM resonator. In the absence of a special design effort to address thisissue, there would be little such over-lap because, as a consequence of a

large difference between wave-lengths, the optical and terahertzmodes would be concentrated at dif-ferent depths from the rim of a WGMresonator. In the present develop-ment, overlap is ensured by con-structing the WGM resonator as aring (see figure) so thin that the opti-cal and terahertz modes are effec-tively forced to overlap.This work was done by Dmitry Strekalov,

Anatoliy Savchenkov, Andrey Matsko, andNan Yu of Caltech for NASA’s Jet Propul-sion Laboratory. Further information is con-tained in a TSP (see page 1).NPO-45508

Determining Concentration of Nanoparticles From EllipsometryCounting of particles is not necessary.Marshall Space Flight Center, Alabama

A method of using ellipsom-etry or polarization analysis oflight in total internal reflec-tion of a surface to determinethe number density of goldnanoparticles on a smoothsubstrate has been developed.The method can be modifiedto enable determination ofdensities of sparse distribu-tions of nanoparticles in gen-eral, and is expected to be es-pecially useful for measuringgold-nanoparticle-labeled bio-molecules on microarrays.

The method is based ontheoretical calculations ofthe ellipsometric responsesof gold nanoparticles. Ele-ments of the calculations in-clude the following:• For simplicity, the gold

nanoparticles are assumedto be spherical and to havethe same radius.

• The distribution of goldnanoparticles is assumedto be a sub-monolayer(that is, sparser than amonolayer).

• The optical response of thesub-monolayer is modeledby use of a thin-island-filmtheory, according to whichthe polarizabilities paralleland perpendicular to thesubstrate are functions ofthe wavelength of light, thedielectric functions (permit-

tivities expressed as complexfunctions of frequency orwavelength) of the gold andthe suspending medium (inthis case, the suspendingmedium is air), the fraction ofthe substrate area covered bythe nanoparticles, and the ra-dius of the nanoparticles.• For the purpose of the

thin-island-film theory, thedielectric function of thegold nanoparticles is mod-eled as the known dielec-tric function of bulk goldplus a correction term thatis necessitated by the factthat the mean free pathlength for electrons in golddecreases with decreasingradius, in such a manner asto cause the imaginary partof the dielectric function toincrease with decreasingradius (see figure). Thecorrection term is a func-tion of the nanoparticle ra-dius, the wavelength oflight, the mean free pathand the Fermi speed ofelectrons in bulk gold, theplasma frequency of gold,and the speed of light in avacuum.These models are used to

calculate ellipsometric re-sponses for various concen-trations of gold nanoparti-cles having an assumed

300 400 500 600 700 800 900 1,000 1,100

300 400 500 600 700 800 900 1,000 1,100

0

2

4

6

8

10

12

-50

-40

-30

-20

-10

0

10

Wavelegth, nm

Imag

inar

y Pa

rtR

eal P

art

5-nm Gold Nanoparticles


Bulk Gold







Bulk Gold

Wavelegth, nm

Real and Imaginary Parts of complex dielectric functions were determinedfor bulk gold and for gold nanoparticles having various diameters.


radius. The modeled data indicatesdistinct spectral features for both thereal and the imaginary part of the di-electric function. An ellipsometricmeasurement would determine thisdistinct feature and thus can be usedto measure nanoparticle concentra-tion. By “ellipsometric responses” ismeant the intensities of light meas-ured in various polarization states as

functions of the angle of incidenceand the polarization states of the inci-dent light. These calculated ellipso-metric responses are used as calibra-tion curves: Data from subsequentellipsometric measurements on realspecimens are compared with the cal-ibration curves. The concentration ofthe nanoparticles on a specimen is as-sumed to be that of the calibration

curve that most closely matches thedata pertaining to that specimen.

This work was done by Srivatsa Venkatasub-barao and Lothar U Kempen of Intelligent Op-tical Systems, Inc. and Russell Chipman of theUniversity of Arizona for Marshall Space FlightCenter. For further information, contact SammyNabors, MSFC Commercialization AssistanceLead, at [email protected]. MFS-32506-1

Microwave-to-Optical Conversion in WGM ResonatorsThree-wave mixing, resonance, and low loss would result in high efficiency.NASA’s Jet Propulsion Laboratory, Pasadena, California

Microwave-to-optical frequency con-verters based on whispering-gallery-mode (WGM) resonators have been pro-posed as mixers for the input ends ofmicrowave receivers in which, down-stream of the input ends, signals wouldbe processed photonically. A frequencyconverter as proposed (see figure)would exploit the nonlinearity of theelectromagnetic response of a WGM res-onator made of LiNbO3 or another suit-able ferroelectric material. Up-conver-sion would take place by three-wavemixing in the resonator.

The WGM resonator would be de -signed and fabricated to obtain (1) res-onance at both the microwave and theoptical operating frequencies and (2)phase matching among the input andoutput microwave and optical signals asdescribed in the immediately precedingarticle. Because the resonator would beall dielectric — there would be nometal electrodes — signal losses wouldbe very low and, consequently, the reso-nance quality factors (Q values) of themicrowave and optical fields would bevery large. The long lifetimes associatedwith the large Q values would enable at-tainment of high efficiency of nonlin-ear interaction with low saturation

power. It is anticipated that efficiencywould be especially well enhanced bythe combination of optical and mi-crowave resonances in operation atinput signal frequencies between 90and 300 GHz.

This work was done by AnatoliySavchenkov, Dmitry Strekalov, Nan Yu,Andrey Matsko, and Lute Maleki of Caltechfor NASA’s Jet Propulsion Laboratory. Fur-ther information is contained in a TSP (seepage 1). NPO-45121

Microwave Waveguide

LiNbO3 WGMResonator

Collimator

OutputOptical Fiber

InputOptical Fiber

CouplingPrism

This Frequency Up-Converter would exploit three-way mixing among a microwave and two opticalsignals.

Four-Pass Coupler for Laser-Diode-Pumped Solid-State LaserA smaller laser slab can be made to perform comparably to a larger one.Goddard Space Flight Center, Greenbelt, Maryland

A four-pass optical coupler affordsincreased (in comparison with relatedprior two-pass optical couplers) utiliza-tion of light generated by a laser diodein side pumping of a solid-state laser

slab. The original application for whichthis coupler was conceived involves aneodymium-doped yttrium aluminumgarnet (Nd:YAG) crystal slab, which,when pumped by a row of laser diodes

at a wavelength of 809 nm, lases at awavelength of 1,064 nm.

Heretofore, typically, a thin laser slabhas been pumped in two passes, thesecond pass occurring by virtue of re-


flection of pump light from a highly re-flective thin film on the side oppositethe side through which the pump lightenters. In two-pass pumping, a Nd:YAGslab having a thickness of 2 mm (whichis typical) absorbs about 84 percent ofthe 809-nm pump light power, leavingabout 16 percent of the pump lightpower to travel back toward the laserdiodes. This unused power can causelocalized heating of the laser diodes,thereby reducing their lifetimes. More-over, if the slab is thinner than 2 mm,then even more unused power travelsback toward the laser diodes.

The four-pass optical coupler capturesmost of this unused pump light andsends it back to the laser slab for twomore passes. As a result, the slab absorbsmore pump light, as though it weretwice as thick. The gain and laser cavitybeam quality of a smaller laser slab inconjunction with this optical couplercan thus be made comparable to thoseof a larger two-pass-pumped laser slab.

The four-pass coupler (see figure)consists of a right-angle polarizationcube (RAPC) with a quarter-wave plateon the side facing the laser slab andhighly reflective film coating one of theperpendicular sides. The RAPC trans-mits p-polarized light (light polarizedparallel to the plane of incidence) andreflects s-polarized light (light polarizedperpendicular to the plane of incidence.Each laser diode emits a collimatedbeam and is oriented so that the beam isp-polarized (vertically polarized in thefigure). The p-polarized beam passesthrough the RAPC, and then throughthe quarter-wave plate, which converts itto a rotationally polarized beam. Thebeam then passes into the laser slab for

a first pump pass, reflection, and secondpump pass in the usual manner.

The pump light remaining after thesecond pass leaves the laser slab andtravels back into the RAPC via the quar-ter-wave plate, which converts this lightto s polarization. This s-polarized beamis reflected from the internal 45° polar-ization beam-splitting surface of theRAPC, sending the beam to the reflec-tive coated RAPC surface at normal in-cidence. After reflection from this sur-face, this beam is reflected by the 45°surface toward the laser slab and is con-verted to rotational polarization by thequarter-wave plate. The beam then

makes two more passes through thelaser slab in the usual manner.

Any pump beam power remainingafter the fourth pass is converted to p po-larization by the quarter-wave plate andtravels back to the laser diode. However,when the coupler is designed correctly inconjunction with the other laser compo-nents, the fraction of pump power re-turning to the laser diode is too small toexert a significant adverse effect on thelaser-diode lifetime or performance.

This work was done by Donald B. Coyle ofGoddard Space Flight Center. Further infor-mation is contained in a TSP (see page 1).GSC-14961-1

The Four-Pass Coupler is interposed between the laser diodes and the laser slab. This view is along thelongitudinal laser axis. The laser diodes, of which only one is shown here, are arranged in a row par-allel to the axis.

Heat Sink

Laser Diode WithCollimating Lens

45° PolarizationBeam-Splitting

Surface

Quarter-Wave Plate

RAPC

LaserSlab

Highly Reflective Surface Films

Low-Resolution Raman-Spectroscopy CombustionThermometry This method offers advantages over related prior Raman-spectroscopy-based methods. John H. Glenn Research Center, Cleveland, Ohio

A method of optical thermometry,now undergoing development, involveslow-resolution measurement of thespectrum of spontaneous Raman scat-tering (SRS) from N2 and O2 mole-cules. The method is especially suitablefor measuring temperatures in high-pressure combustion environmentsthat contain N2, O2, or N2/O2 mixtures(including air).

Methods based on SRS (in which scat-tered light is shifted in wavelength byamounts that depend on vibrational androtational energy levels of laser-illumi-nated molecules) have been popularmeans of probing flames because theyare almost the only methods that pro-vide spatially and temporally resolvedconcentrations and temperatures ofmultiple molecular species in turbulent

combustion. The present SRS-basedmethod differs from prior SRS-basedmethods that have various drawbacks, adescription of which would exceed thescope of this article. Two main differ-ences between this and prior SRS-basedmethods are that • It involves analysis in the frequency

(equivalently, wavelength) domain, incontradistinction to analysis in the in-


tensity domain in prior methods; and • It involves low-resolution measure-

ment of what amounts to predomi-nantly the rotational Raman spectra ofN2 and O2, in contradistinction tohigher-resolution measurement of thevibrational Raman spectrum of N2 onlyin prior methods.Analysis in the frequency domain re-

duces the effects of uncertainties in thespectral-response calibration and per-mits greater signal-to-noise ratios by ex-cluding the noise contributed by inten-sity or amplitude fluctuations. Oneadvantage of utilizing the rotationalRaman spectral bands is that they aremuch stronger than are the vibrationalRaman spectral bands. In particular, inthis method, one utilizes the rotationalN2 bands near the laser wavelength. Thedeliberate choice of lower resolutionmakes it acceptable to use wider spectro-graph slits and thereby to collect morelight to obtain greater signal-to-noise ra-tios. A further advantage of lower reso-lution is the independence of the spec-tra on pressure broadening effects.

According to theoretical simulations,the rotational Raman spectrum of N2

widens with increasing temperature (seeFigure 1). This is because at higher tem-perature, greater proportions of rota-tional states having higher energies be-come excited. Consequently, it shouldbe possible to establish a relationship be-tween the width Wd of the envelope ofthe rotational Raman spectrum and thetemperature and to express this relation-ship as a conversion formula for deter-mining the temperature from Wd of ameasured spectrum; this is the basicprinciple of the present method. Themethod as described thus far would besimple, were it not for the facts that (1)the rotational Raman spectra of N2 andO2 overlap and (2) almost any practicalcombustion system contains N2 and O2.The net effect of the superposition ofthe N2 and O2 rotational Raman spectrais to produce a taller, narrower versionof the spectrum of pure N2, the amountof narrowing depending on the relativeproportions of N2 and O2.

To account for this narrowing, it be-comes necessary to generate and use amore comprehensive conversion for-mula, as illustrated in Figure 2. First, theenvelopes of rotational SRS spectra of N2

and O2 are calculated theoretically over arange of temperature at a certain pres-sure to obtain the conversion formulasfor N2 and pure O2. Then a blended con-version formula is obtained as a weightedaverage, wherein the weighting factors

Figure 1. Rotational Raman Spectra of pure N2 at three temperatures were calculated to show howWd increases with temperature.

0

0

0

0

0

0

Wd

Wd

Wd

Wavelength, nm Wavelength, nm

510 520 530 540 550 510 520 530 540 550

510 520 530 540 550 510 520 530 540 550

510 520 530 540 550 510 520 530 540 550

Temperature = 500 K Temperature = 500 K

Temperature = 1,500 K Temperature = 1,500 K

Temperature = 2,500 K Temperature = 2,500 K

Rel

ativ

e In

ten

sity

Rel

ativ

e In

ten

sity

Rel

ativ

e In

ten

sity

High-Resolution PlotsShowing Spectral Lines

Low-Resolution PlotsShowing Envelopes

Figure 2. The Conversion Formulas for Pure N2 and O2 are represented by the lower and upper bound-aries, respectively, of the shaded region. For an N2/O2 mixture, the conversion formula lies betweenthese boundaries.

5 7 9 11 13 15

Tem

per

atu

re, K

Wd: Rotational Raman bandwidth, nm

3,000

2,500

2,000

1,500

1,000

500 Wd

Pure O2

Pure N2

Mixture of 7.5 Parts N2 and1 Part O2


Temperature Sensors Based on WGM Optical ResonatorsDifferences between temperature-dependent frequencies of resonances would be measured.NASA’s Jet Propulsion Laboratory, Pasadena, California

A proposed technique for measuringtemperature would exploit differencesbetween the temperature dependencesof the frequencies of two different elec-tromagnetic modes of a whispering-gallery-mode (WGM) optical resonator.An apparatus based on this techniquewas originally intended to be part of acontrol system for stabilizing a laserfrequency in the face of temperaturefluctuations. When suitably calibrated,apparatuses based on this techniquecould also serve as precise temperaturesensors for purposes other than stabi-lization of lasers.

A sensor according to the proposalwould include (1) a transparent WGMdielectric resonator having at least twodifferent sets of modes characterized bydifferent thermo-optical constants and(2) optoelectronic instrumentation formeasuring the difference between thetemperature-dependent shifts of the res-

onance frequencies of the two sets ofmodes. The figure schematically depictsan example of such a sensor. Laser 1, op-erating at frequency f0, would be lockedto a mode in the first of the two sets ofWGM modes to be exploited; the modelocking would be accomplished by estab-lished means that would include pho-todetector 1, an oscillator, polarizers,mixer 1, and electro-optical modulatorEOM 1. Laser 2, operating at frequency2f0 + δf, would be locked to a mode in the second of the two sets ofWGM modes to be exploited; in thiscase, the mode locking would be accom-plished by established means that wouldinclude photodetector 2, the oscillator,mixer 1, and electro-optical modulatorEOM 2.

Part of the modulated output of laser1 would be fed through a frequency dou-bler to obtain a modulated beam at fre-quency 2f0. In a beam splitter, the 2f0

output from the frequency doublerwould be combined with part of themodulated output of laser 2 at 2f0 + δf.The interference between these com-bined beams would cause the output ofphotodetector 3 to include a compo-nent at the heterodyne frequency, δf,which would have the desired tempera-ture dependence. Inasmuch as f0 and δfcould readily be chosen to place δfwithin a suitable radio-frequency rangeand means for measuring radio fre-quency precisely are readily available, itwould be straightforward to measure δf.Then the temperature could be calcu-lated by inversion of the known temper-ature dependence of δf. It has been esti-mated that for a typical CaF2 WGMresonator having a resonance quality fac-tor (Q) of 2 × 1010, the temperature-measurement sensitivity would be char-acterized by a temperature increment ofabout 40 µK for a frequency increment

are determined by the relative propor-tions of N2 and O2 as measured or calcu-lated by independent means. (The inde-pendent means could be measurementsof vibrational Raman spectra of N2 andO2 or a chemical-equilibrium calcula-

tion.) Finally, the measured Wd is insertedinto the blended conversion formula,yielding the temperature.

This work was done by Quang-Viet Nguyenof Glenn Research Center and Jun Kojima ofOhio Aerospace Institute.

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-18100-1.

Laser 1,Frequency f0 EOM 1

EOM 2

Polarizers

WGMResonator

Photodetector 1

Photodetector 2

Photodetector 3

Mixer 1

Mixer 2

FrequencyDoubler

BeamSplitter

OscillatorElectronicConnection

OpticalConnection

Polarizers

The Lasers Would Be Mode-Locked to resonance frequencies f0 and 2f0 + δf, respectively, of the WGM resonator. The heterodyne frequency, δf, would varywith temperature and, therefore, would be measured for use as an indication of temperature.


of half the width of one of the resonancespectral peaks.

This work was done by AnatoliySavchenkov, Nan Yu, Lute Maleki, VladimirIltchenko, Andrey Matsko, and Dmitry

Strekalov of Caltech for NASA’s Jet PropulsionLaboratory. Further information is containedin a TSP (see page 1).

This invention is owned by NASA, anda patent application has been filed. In-

quiries concerning nonexclusive or exclu-sive license for its commercial developmentshould be addressed to the Patent Counsel,NASA Management Office–JPL. Refer toNPO-44469.

Varying the Divergence of Multiple Parallel Laser BeamsLenses mode-matched to the laser beams would be moved axially within an afocal opticalsubassembly.NASA’s Jet Propulsion Laboratory, Pasadena, California

A provision for controlled variation ofthe divergence of a laser beam or of mul-tiple parallel laser beams has been incor-porated into the design of a conceptualfree-space optical-communication stationfrom which the transmitted laser beam(s)would be launched via a telescope. Theoriginal purpose to be served by this pro-vision was to enable optimization, undervarious atmospheric optical conditions,of the divergence of a laser beam orbeams transmitted from a ground stationto a spacecraft. Beyond the original pur-pose, the underlying design conceptcould be beneficial for terrestrial free-space laser communication, ranging, andscientific instrumentation applications inwhich there are requirements to vary thedivergences of laser beams.

In order to be able to provide for con-trolled variation of beam divergence,one must first gain detailed understand-ing of the optical train from each laserto the primary mirror of the telescope.Gaussian propagation of each laser

beam through all the optical elementsmust be computed. If multiple parallelbeams were to be transmitted, then bymeans of previously developed optics,they would be positioned symmetricallyabout the optical axis. It would be neces-sary to perform paraxial ray tracing toensure that the beams emerging fromthe primary mirror into free space wereparallel to each other and to the mainoptical axis of the telescope.

The design concept reflects a re-quirement in the original applicationthat final divergence of the beam(s)propagating out from the primary mir-ror into free space be varied by movingonly one lens or lens assembly in theoptical train and that this motion notcause the outgoing beam(s) to deviatefrom parallelism with the optical axis.To satisfy this requirement, the tele-scope would incorporate an afocal op-tical subassembly, within which eithera single on-axis lens in the case of asingle laser beam or a ring assembly of

lenses in the case of multiple laserbeams (see figure) would be moved.The lens or lenses must be designed tomode-match the laser output throughthe afocal subassembly and telescopeoptics to produce the required beamdivergence. By moving this lens (ormoving the assembly of lenses as a sin-gle unit) along the optical axis, onewould cause the divergence of the out-going laser beam(s) to vary throughthe required range. Care must betaken to ensure that there is noapodization or vignetting through anylimiting apertures in the overall opti-cal system and that power density ofany laser beam must not be so high asto result in dielectric breakdown of airor in damage to any optic along theoptical path.

This work was done by Joseph M. Kovalikand Malcolm W. Wright of Caltech forNASA’s Jet Propulsion Laboratory. For moreinformation, contact [email protected]

AfocalRegion

LaserBeams Frame Holding

Two Lenses

LensTelescope

Focus

PrimaryTelescope Mirror

OutputBeams

Optical Axis

A Frame Would Hold Lenses placed symmetrically about the optical axis to intercept multiple laser beams parallel to the axis. (For simplicity, only twobeams are shown here, but the original design calls for eight beams). The frame would be moved along the optical axis to vary the divergence of thelaser beams emerging from the primary telescope mirror. The motion of the frame and lenses would not cause the beams to deviate from parallelismwith the optical axis.


Information Sciences

Algorithm-Based Fault Tolerance Integrated With ReplicationNASA’s Jet Propulsion Laboratory, Pasadena, California

In a proposed approach to program-ming and utilization of commercialoff-the-shelf computing equipment, acombination of algorithm-based faulttolerance (ABFT) and replicationwould be utilized to obtain high de-grees of fault tolerance without incur-ring excessive costs. The basic idea ofthe proposed approach is to integrateABFT with replication such that the al-gorithmic portions of computations

would be protected by ABFT, and thelogical portions by replication.

ABFT is an extremely efficient, inex-pensive, high-coverage technique for de-tecting and mitigating faults in com-puter systems used for algorithmiccomputations, but does not protectagainst errors in logical operations sur-rounding algorithms. Replication is agenerally applicable, high-coveragetechnique for protecting general com-

putations from faults, but is inefficientand costly because it requires additionalcomputation time or additional compu-tational circuitry (and, hence, additionalmass and power). The goal of the pro-posed integration of ABFT with replica-tion is to optimize the fault-tolerance as-pect of the design of a computing systemby using the less-efficient, more-expen-sive technique to protect only thosecomputations that cannot be protected

Efficient Algorithm for Rectangular Spiral SearchThe search pattern is automatically expanded as needed.NASA’s Jet Propulsion Laboratory, Pasadena, California

An algorithm generates grid coordi-nates for a computationally efficient spi-ral search pattern covering an uncertainrectangular area spanned by a coordi-nate grid. The algorithm does not re-quire that the grid be fixed; the algo-rithm can search indefinitely, expandingthe grid and spiral, as needed, until thetarget of the search is found. The algo-rithm also does not require memory ofcoordinates of previous points on thespiral to generate the current point onthe spiral.

The search is started at a point (moreprecisely, in a grid cell), denoted thecenter of the spiral, that has been cho-sen previously as the point most likely tocoincide with the target. The search is tobe performed on a grid of n×m cells,where m>n and a ≡ m – n is denoted the

rectangular excess of the search pattern(see Figure 1). The spiral search is per-formed in steps numbered simply 1, 2,3..., and t represents the number of thecurrent step.

The inputs to the algorithm are tand the rectangular excess that is ei-ther specified in advance or calculatedfrom the rectangular grid used to spanthe initial search area. The output ofthe algorithm is the pair of integer co-ordinates (i,j) of the current point onthe spiral with respect to the center of

the spiral. Figure 2 presents, as an ex-ample, the first 15 steps of the spiralgenerated by this algorithm for asearch that starts at a point in a 3×5rectangle.

This work was done by Paul Brugarolasand William Breckenridge of Caltech forNASA’s Jet Propulsion Laboratory.

The software used in this innovation isavailable for commercial licensing. Pleasecontact Karina Edmonds of the California In-stitute of Technology at (626) 395-2322.Refer to NPO-42057.

m

n

1 2 3 4

56789

10

11 12 13 14 15

Figure 2. A Spiral Search Pattern was generated in an area initially overlaid with a 3×5 grid.Figure 1. A Rectangular Grid of n×m cells is over-laid on the area to be searched.


Targeting and Localization for Mars Rover OperationsNASA’s Jet Propulsion Laboratory, Pasadena, California

A design and a partially developed ap-plication framework were presented forimproving localization and targeting forsurface spacecraft. The program hasvalue for the Mars Science Laboratorymission, and has been delivered to sup-port the Mars Exploration Rovers as partof the latest version of the Maestro sci-ence planning tool. It also has applica-tions for future missions involving eithersurface-based or low-altitude atmos-pheric robotic vehicles.

The targeting and localization solu-tions solve the problem of how to inte-grate localization estimate updates into

operational planning tools, operationaldata product generalizations, and flightsoftware by adding expanded flexibilityto flight software, the operations dataproduct pipeline, and operations plan-ning tools based on coordinate frameupdates during a planning cycle. Whenacquiring points of interest (targets)for the rover, instead of using a tempo-ral method for reusing previously ac-quired targets, this system uses a spatialmethod to avoid tedious and repetitivetarget re-designation needed to keeptarget relevance accurate. Instead ofcreating a target that is reusable only

for a sol (Martian day), the target is de-fined in a way to make it reusable for aplanning position (the vehicle positionindicated by a Site and Drive indexpair) from which the vehicle will begina command cycle.

This work was done by Mark W. Powell,Thomas Crockett, Jason M. Fox, Joseph C.Joswig, Jeffrey S. Norris, and Kenneth J.Rabe of NASA’s Jet Propulsion Laboratory.

The software used in this innovation isavailable for commercial licensing. Pleasecontact Karina Edmonds of the CaliforniaInstitute of Technology at (626) 395-2322.Refer to NPO-43847.

Terrain-Adaptive Navigation Architecture NASA’s Jet Propulsion Laboratory, Pasadena, California

A navigation system designed for a Marsrover has been designed to deal withrough terrain and/or potential slip whenevaluating and executing paths. The sys-tem also can be used for any off-road, au-tonomous vehicles. The system uses moresophisticated terrain analysis, but also con-verges to computational complexity simi-lar to that of currently deployed naviga-tion systems when the terrain is benign.The system consists of technologies thathave been developed, integrated, and

tested onboard research rovers in Marsanalog terrains, including goodness mapsand terrain triage, terrain classification, re-mote slip prediction, path planning, high-fidelity traversability analysis (HFTA), andslip-compensated path following.

The system enables vehicles to auto -nomously navigate different terrainchallenges including dry river channelsystems, putative shorelines, and gulliesemanating from canyon walls. Several ofthe technologies within this innovation

increase the navigation system’s capabil-ities compared to earlier rover naviga-tion algorithms.

This work was done by Daniel M. Helmick,Anelia Angelova, Larry H. Matthies, andDaniel M. Helmick of Caltech for NASA’s JetPropulsion Laboratory.

The software used in this innovation isavailable for commercial licensing. Pleasecontact Karina Edmonds of the California In-stitute of Technology at (626) 395-2322.Refer to NPO- 44588.

by the more-efficient, less-expensivetechnique. It would not be necessary toaddress the fault-tolerance issue explic-itly in writing an application program tobe executed in such a system. Instead,

ABFT and replication would be man-aged by middleware containing hooks.

This work was done by Raphael Some andDavid Rennels of Caltech for NASA’s JetPropulsion Laboratory.

The software used in this innovation isavailable for commercial licensing. Pleasecontact Karina Edmonds of the CaliforniaInstitute of Technology at (626) 395-2322.Refer to NPO-43842.

A common practice in computer sci-ence to associate a value with a key is touse a class of algorithms called a hash-table. These algorithms enable rapidstorage and retrieval of values basedupon a key. This approach assumesthat many keys will need to be storedimmediately. A new set of hash-table al-gorithms optimally uses system re-sources to ideally represent keys and

values in memory such that the infor-mation can be stored and retrievedwith a minimal amount of time andspace. These hash-tables support theefficient addition of new entries. Also,for large data sets, the look-up time forlarge data-set searches is independentof the number of items stored, i.e.,O(1), provided that the chance of col-lision is low.

Like arrays, hash-tables provide con-stant time O(1) look-up on average, re-gardless of the number of items in thetable. However, the rare worst-case look-up time can be as bad as O(n). Com-pared to other associative array datastructures, hash-tables are most usefulwhen large numbers of records are to bestored, especially if the size of the dataset can be predicted.

Self-Adjusting Hash Tables for Embedded Flight ApplicationsNASA’s Jet Propulsion Laboratory, Pasadena, California


An Extensible Markup Language(XML) schema was developed as ameans of defining and describing astructure for capturing spacecraft com-mand-definition and tracking informa-tion in a single location in a form read-able by both engineers and softwareused to generate software for flight andground systems. A structure definedwithin this schema is then used as thebasis for creating an XML file that con-tains command definitions. The schemais divided into three sections:• Header information, including infor-

mation about the project and XML fileto be derived from the schema;

• Project-specific definitions of types,roles, and allowable values of data; and

• The information necessary for defin-ing the command structure, includingthe information necessary for generat-ing all pertinent software.Among the advantages afforded by

XML for such applications are the fol-lowing:• There exist commercial off-the-shelf

(COTS) software tools and standardscripting-language modules for pars-

ing XML schemata. These tools andmodules facilitate the ingestion ofXML files for use.

• By use of COTS software tools, thestructures of, and some properties ofthe data in, XML files can be validatedagainst their parent XML schemata todetect errors early.This work was done by Sharon Laubach,

Celina Garcia, Scott Maxwell, and JesseWright of Caltech for NASA’s Jet PropulsionLaboratory. For more information, [email protected]

Schema for Spacecraft-Command DictionaryNASA’s Jet Propulsion Laboratory, Pasadena, California

These algorithms may be used as in-memory data structures and may also beadopted for use with persistent datastructures. Hash-tables are used in thou-sands of instances of flight code, and thisapproach could improve the efficiencyof those applications, allowing them torun in smaller memory footprints and toautonomously evolve with time if and

when their data demands a more effi-cient representation.

This work was done by Mark James of Cal-tech for NASA’s Jet Propulsion Laboratory.Further information is contained in a TSP(see page 1).

In accordance with Public Law 96-517,the contractor has elected to retain title to thisinvention. Inquiries concerning rights for its

commercial use should be addressed to:Innovative Technology Assets ManagementJPLMail Stop 202-2334800 Oak Grove DrivePasadena, CA 91109-8099E-mail: [email protected] to NPO-40363, volume and number of this

NASA Tech Briefs issue, and the page number.


Books & Reports

Combined GMSK Communi-cations and PN Ranging

A document discusses a method bywhich GMSK (Gaussian minimum shiftkeying) modulation and a pseudonoise(PN) ranging signal may be combined.By isolating the in-phase and quadraturecomponents after carrier lock, and ex-tracting their low-pass and band-pass fil-tered components, there is enough in-formation available to both demodulatedata and track the PN signal. The pro-posed combined GMSK communica-tions and PN ranging is one potentialapproach to address emerging require-ments for simultaneous high data ratecommunications from and tracking ofvehicles in deep space or at the Moon.

GMSK and PN ranging have not beenpreviously combined, and the correspon-ding receiver structure for such combinedranging has not been proposed in thepast. A key advantage is that the com-bined signal is bandwidth-efficient and itis a constant envelope modulation, allow-ing high-power amplifiers to operate atsaturation for highest efficiency.

This work was done by Richard Orr ofSATEL LLC and Dariush Divsalar of Caltechfor NASA’s Jet Propulsion Laboratory. Furtherinformation is contained in a TSP (see page1). NPO-45108

System-Level Integration ofMass Memory

A report discusses integrating multiplememory modules on the high-speed serialinterconnect (IEEE 1393) that is used bya spacecraft’s inter-module communica-tions in order to ease data congestion andprovide for a scalable, strong, flexible sys-tem that can meet new system-level massmemory requirements.

Using the JPL 1393 Ring Bus Inter-connect to link computer elements,I/O, and memory allows any element tocommunicate with any other element.Besides providing a consistent approachto exchanging data, it inherently has alayer of abstraction that allows for bettersystem and software design. This new ar-chitecture is fault-tolerant and providesa large range of scalability while support-ing flexible spacecraft architectures thatare currently being investigated.

This work was done by Brian Cox, JeffreyMellstrom, and Terry Wysocky of Caltech

for NASA’s Jet Propulsion Laboratory. Further information is contained in a TSP (seepage 1).


Innovative Technology Assets ManagementJPLMail Stop 202-2334800 Oak Grove DrivePasadena, CA 91109-8099E-mail: [email protected] to NPO-45205, volume and number


Network-Attached Solid-StateRecorder Architecture

A document discusses placing mem-ory modules on the high-speed serial in-terconnect, which is used by a space-craft’s computer elements forinter-processor communications, toallow all multiple computer system ar-chitectures to access the spacecraft datastorage at the same time. Each memoryboard is identical electrically and re-ceives its bus ID upon connection to thesystem. The computer elements are con-figured in a similar fashion. The archi-tecture allows for multiple memoryboards to be accessed simultaneously bydifferent computer elements, and re-sults in a scalable, strong, fault-tolerantsystem. The IEEE-1393 ring bus can berouted so that multiple card failures canoccur and the mass memory storage willstill function.

This work was done by Brian Cox of Caltechfor NASA’s Jet Propulsion Laboratory. Furtherinformation is contained in a TSP (see page1). NPO-45204

Method of Cross-LinkingAerogels Using a One-PotReaction Scheme

A document discusses a new, simpli-fied method for cross-linking silica andother oxide aerogels, with a polymericmaterial to increase strength of such ma-terials without adversely affecting poros-ity or low density. The usual process islong and arduous, requiring multiplewashing and soaking steps to infiltrateoxide with the polymer precursor aftergelation. Additionally, diffusion prob-

lems can result in aerogel monolithsthat are not uniformly cross-linked.

This innovation introduces the poly-mer precursor into the sol before gela-tion either as an agent, which co-reactswith the oxide gel, or as soluble polymerprecursors, which do not interact withthe oxide gel in any way. Subsequent ex-posure to heat, light, catalyst or othermethod of promoting polymerizationcauses cross-linking without any addi-tional infiltration steps.

The resulting aerogel monolith ismore uniform because this process doesnot suffer from diffusion issues that pre-vious methods have. Also, in instanceswhere complete polymerization requiresa balanced stoichiometry, this require-ment is more easily met.

This work was done by Mary Ann B.Meador and Lynn A. Capadona 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, InnovativePartnerships Office, Attn: Steve Fedor, MailStop 4–8, 21000 Brookpark Road, Cleve-land, Ohio 44135. Refer to LEW-18042.

An Efficient ReachabilityAnalysis Algorithm

A document discusses a new algo-rithm for generating higher-order de-pendencies for diagnostic and sensorplacement analysis when a system is de-scribed with a causal modeling frame-work. This innovation will be used in di-agnostic and sensor optimization andanalysis tools. Fault detection, diagno-sis, and prognosis are essential tasks inthe operation of autonomous space-craft, instruments, and in-situ plat-forms. This algorithm will serve as apower tool for technologies that satisfya key requirement of autonomousspacecraft, including science instru-ments and in-situ missions.

In the causal modeling, the system ismodeled in terms of first-order cause-and- effect dependencies; i.e., how thefault propagates from a faulty compo-nent to its immediate neighbors. For di-agnostic purpose, also global (or higher-order) dependencies are needed, whichis the effect of a fault on non-neighborcomponents. The global dependenciesshould be inferred from the first-order


dependencies. The method that findsthese dependencies is called a reachabil-ity analysis algorithm. The result of thisalgorithm determines at each test point(or sensor position) which of the failuresources can be observed.

The standard reachability analysis al-gorithm uses a “token propagation”method. The complexity of this algo-

rithm is proportional to the product EN,where E is the number of links (edges)of the graph of the system and N is thenumber of components. Here a new al-gorithm is introduced. The complexityof this algorithm is proportional to theproduct dN, where d is the length of thelongest (directed) path in the graph ofthe system. To compare the perform-

ance of these two algorithms, first it isnoted that always d ≤ E. But typically, d isof the order of log(E); thus the new al-gorithm, in general, outperforms thestandard algorithm.

This work was done by Farrokh Vatan andAmir Fijany of Caltech for NASA’s Jet Propul-sion Laboratory. Further information is con-tained in a TSP (see page 1). NPO-45797

National Aeronautics andSpace Administration

Documents

Technology Focus Electronics/Computers · 2013-04-10 · NASA Tech Briefs, November 2008 1 INTRODUCTION Tech Briefs are short announcements of innovations originating from research