technology focus electronics/computers - nasa€¦ · tripler at 350-mw input power, and 27-mw...

Technology Focus

Electronics/Computers

Software

Materials

Mechanics/Machinery

Manufacturing

Bio-Medical

Physical Sciences

Information Sciences

Books and Reports

12-10 December 2010

https://ntrs.nasa.gov/search.jsp?R=20100042204 2020-06-22T14:28:39+00:00Z

NASA Tech Briefs, December 2010 1

INTRODUCTIONTech Briefs are short announcements of innovations originating from research and developmentactivities of the National Aeronautics and Space Administration. They emphasize information con-sidered likely to be transferable across industrial, regional, or disciplinary lines and are issued toencourage 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 www.nttc.edu

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://www.nasa.gov/offices/ipp/network/index.html

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 CenterYvonne D. Gibbs(661) [email protected]

Glenn Research CenterKathy Needham(216) [email protected]

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

Jet Propulsion LaboratoryAndrew Gray(818) [email protected]

Johnson Space Centerinformation(281) [email protected]

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

Langley Research CenterElizabeth B. Plentovich(757) [email protected]

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

Stennis Space CenterRamona Travis(228) 688-3832 [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]

5 Technology Focus: Electronic Components

5 Coherent Frequency Reference System for theNASA Deep Space Network

5 Diamond Heat-Spreader for Submillimeter-WaveFrequency Multipliers

6 180-GHz I-Q Second Harmonic Resistive Mixer MMIC

6 Ultra-Low-Noise W-Band MMIC Detector Modules

6 338-GHz Semiconductor Amplifier Module

6 Power Amplifier Module With 734-mWContinuous Wave Output Power

9 Electronics/Computers9 Multiple Differential-Amplifier MMICs

Embedded in Waveguides

9 Rapid Corner Detection Using FPGAs

10 Special Component Designs for Differential-Amplifier MMICs

10 Multi-Stage System for Automatic Target Recognition

11 Single-Receiver GPS Phase Bias Resolution

12 Ultra-Wideband Angle-of-Arrival Tracking Systems

13 Update on Waveguide-Embedded DifferentialMMIC Amplifiers

15 Software15 Automation Framework for Flight Dynamics

Products Generation

15 Product Operations Status Summary Metrics

15 Mars Terrain Generation

15 Application-Controlled Parallel AsynchronousInput/Output Utility

16 Planetary Image Geometry Library

17 Manufacturing & Prototyping17 Propulsion Design With Freeform

Fabrication (PDFF)

17 Economical Fabrication of Thick-Section Ceramic Matrix Composites

18 Process for Making a Noble Metal on Tin Oxide Catalyst

19 Mechanics/Machinery19 Stacked Corrugated Horn Rings

19 Refinements in an Mg/MgH2/H2O-BasedHydrogen Generator

20 Continuous/Batch Mg/MgH2/H2O-BasedHydrogen Generator

21 Strain System for the Motion Base ShuttleMission Simulator

21 Ko Displacement Theory for Structural sShape Predictions

22 Pyrotechnic Actuator for Retracting TubesBetween MSL Subsystems

23 Physical Sciences23 Surface-Enhanced X-Ray Fluorescence

24 Infrared Sensor on Unmanned Aircraft TransmitsTime-Critical Wildfire Data

24 Slopes To Prevent Trapping of Bubbles inMicrofluidic Channels


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

12-10 December 2010


Technology Focus: Electronic Components

The NASA Deep Space Network(DSN) requires state-of-the-art frequencyreferences that are derived and distrib-uted from very stable atomic frequencystandards. A new Frequency ReferenceSystem (FRS) and Frequency ReferenceDistribution System (FRD) have been de-veloped, which together replace the pre-vious Coherent Reference Generator Sys-tem (CRG). The FRS and FRD eachprovide new capabilities that significantlyimprove operability and reliability.

The FRS allows for selection andswitching between frequency standards,a flywheel capability (to avoid interrup-tions when switching frequency stan-

dards), and a frequency synthesis system(to generate standardized 5-, 10-, and100-MHz reference signals). The FRS ispowered by redundant, specially filtered,and sustainable power systems and in-cludes a monitor and control capabilityfor station operations to interact andcontrol the frequency-standard selectionprocess. The FRD receives the standard-ized 5-, 10-, and 100-MHz reference sig-nals and distributes signals to distribu-tion amplifiers in a fan out fashion todozens of DSN users that require thehighly stable reference signals. The FRDis also powered by redundant, speciallyfiltered, and sustainable power systems.

The new DSN Frequency DistributionSystem, which consists of the FRS andFRD systems described here, is central toall operational activities of the NASADSN. The frequency generation and dis-tribution system provides ultra-stable, co-herent, and very low phase-noise refer-ences at 5, l0, and 100 MHz to between60 and 100 separate users at each DeepSpace Communications Complex.

This work was done by Blake C. Tucker,John E. Lauf, Robert L. Hamell, Jorge Gonza-lez, Jr., William A. Diener, and Robert L.Tjoelker of Caltech for NASA’s Jet PropulsionLaboratory. For more information, [email protected]. NPO-46602

Coherent Frequency Reference System for the NASA DeepSpace NetworkNASA’s Jet Propulsion Laboratory, Pasadena, California

The planar GaAs Shottky diode fre-quency multiplier is a critical technologyfor the local oscillator (LO) for submil-limeter-wave heterodyne receivers dueto low mass, tenability, long lifetime, androom-temperature operation. The useof a W-band (75–100 GHz) power ampli-fier followed by a frequency multiplier isthe most common for submillimeter-wave sources. Its greatest challenge is toprovide enough input power to the LOfor instruments onboard future plane-tary missions.

Recently, JPL produced 800 mW at92.5 GHz by combining four MMICs inparallel in a balanced configuration. Asmore power at W-band is available to themultipliers, their power-handling capa-bility be comes more important. Highoperating temp eratures can lead todegradation of conversion efficiency orcatastrophic failure.

The goal of this innovation is to re-duce the thermal resistance by attachingdiamond film as a heat-spreader on thebackside of multipliers to improve their

power-handling capability. Polycrys-talline diamond is deposited by hot-fila-ment chemical vapor deposition (CVD).This diamond film acts as a heat-spreader to both the existing 250- and300-GHz triplers, and has a high thermalconductivity (1,000–1,200 W/mK). It isapproximately 2.5 times greater thancopper (401 W/mK) and 20 timesgreater than GaAs (46 W/mK). It is anelectrical insulator (resistivity ≈1015 Ω·cm), and has a low relative dielectricconstant of 5.7.

Diamond heat-spreaders reduce by atleast 200 ºC at 250 mW of input power,compared to the tripler without dia-mond, according to thermal simulation.This superior thermal management pro-vides a 100-percent increase in power-handling capability. For example, withthis innovation, 40-mW output powerhas been achieved from a 250-GHztripler at 350-mW input power, while theprevious triplers, without diamond, suf-fered catastrophic failures. This break-through provides a stepping-stone for

frequency multipliers-based LO up to 3THz. The future work for this design isto apply the high output power fromboth the 250 and 300 GHz to multiplechains in order to generate milliwatts at2–3 THz.

Using the first generation of resultsfor this innovation, 40 mW of outputpower were produced from a 240-GHztripler at 350-mW input power, and 27-mW output power was produced from a300-GHz tripler at 408-mW input power.This is two times higher than the currentstate-of-the-art output power capability.A finite-element thermal simulation alsoshows that 30-µm thick diamonddropped the temperature of the anodesby at least 200 ºC.

This work was done by Robert H. Lin, Erich T.Schlecht, Goutam Chattopadhyay, John J. Gill,Imran Mehdi, Peter H. Siegel, John S. Ward,Choonsup Lee, and Bertrand C. Thomas of Cal-tech and Alain Maestrini of University Pierre etMarie Curie Paris for NASA’s Jet Propulsion Lab-oratory. For more information, [email protected]. NPO-46777.

Diamond Heat-Spreader for Submillimeter-Wave FrequencyMultipliers Superior thermal management provides a 100-percent increase in power-handling capability. NASA’s Jet Propulsion Laboratory, Pasadena, California

6 NASA Tech Briefs, December 2010

Research findings were reported froman investigation of new gallium nitride(GaN) monolithic millimeter-wave inte-grated circuit (MMIC) power amplifiers

(PAs) targeting the highest output powerand the highest efficiency for class-A op-eration in W-band (75–110 GHz). W-bandPAs are a major component of many fre-

quency multiplied submillimeter-wave LOsignal sources. For spectrometer arrays,substantial W-band power is required dueto the passive lossy frequency multipliers

Power Amplifier Module With 734-mW Continuous WaveOutput Power NASA’s Jet Propulsion Laboratory, Pasadena, California

A monolithic microwave integratedcircuit (MMIC) receiver can be used asa building block for next-generationradio astronomy instruments that arescalable to hundreds or thousands ofpixels. W-band (75–110 GHz) low-noisereceivers are needed for radio astron-omy interferometers and spectrome-ters, and can be used in missile radarand security imagers.

These receivers need to be designed tobe mass-producible to increase the sensi-tivity of the instrument. This innovation isa prototyped single-sideband MMIC re-ceiver that has all the receiver front-endfunctionality in one small and planarmodule. The planar module is easy to as-semble in volume and does not requiretuning of individual receivers. This makesthis design low-cost in large volumes.

This work was done by Todd C. Gaier,Lorene A. Samoska, and Pekka P. Kan-gaslahti of Caltech; Dan Van Vinkle, SamiTantawi, John Fox, Sarah E. Church, JudyM. Lau, Matthew M. Sieth, and Patricia E.Voll of Stanford University; and Eric Bryertonof NRAO for NASA’s Jet Propulsion Labora-tory. Further information is contained in aTSP (see page 1). NPO-47348

A 35-nm-gate-length InP, high-elec-tron-mobility transistor (HEMT) with ahigh-indium-content channel as the keycomponent was developed to producean MMIC (monolithic microwave inte-grated circuit) power amplifier. With ashorter gate length than previous tran-sistor generations, it allows for elec-trons to travel shorter distances. Thisresults in higher frequency functional-ity. In addition, the fabrication processprovides for a comprehensive passivecomponent library of resistors, capaci-

tors, airbridge wiring, and through-wafer vias that allow for transistor RFmatching and power combining on-chip, making the measured 10-mW 338-GHz chip possible.

The amplifier module can be used inseries with current ≈340 GHz RF sourcesto boost RF output power. The ex-tremely high-frequency power amplifiermodule can be used for very-high-fre-quency wide-bandwidth communica-tion, and higher resolution radars forcivilian applications.

This work was done by Lorene A.Samoska, Todd C. Gaier, Mary M. Soria,and King Man Fung of Caltech and VesnaRadisic, William Deal, Kevin Leong, XiaoBing Mei, Wayne Yoshida, Po-Hsin Liu,Jansen Uyeda, and Richard Lai of NorthropGrumman Corp. for NASA’s Jet PropulsionLaboratory.

The software used in this innovation isavailable for commercial licensing. Pleasecontact Daniel Broderick of the California In-stitute of Technology at [email protected] to NPO-47307.

338-GHz Semiconductor Amplifier ModuleNASA’s Jet Propulsion Laboratory, Pasadena, California

Ultra-Low-Noise W-Band MMIC Detector Modules NASA’s Jet Propulsion Laboratory, Pasadena, California

An indium phosphide MMIC (mono-lithic microwave integrated circuit) mixerwas developed, processed, and tested inthe NGC 35-nm-gate-length HEMT (highelectron mobility transistor) process. TheMMIC mixers were tested and assembledin the miniature MMIC receiver moduledescribed in “Miniature Low-Noise G-Band

I-Q Receiver” (NPO-47442), NASA TechBriefs, Vol. 34, No. 11 (November 2010), p.45. This innovation is very compact in sizeand operates with very low LO power. Be-cause it is a resistive mixer, this innovationdoes not require DC power. This is an en-abling technology for the miniature re-ceiver modules for the GeoSTAR instru-

ment, which is the only viable option forthe NRC decadal study mission PATH.

This work was done by Pekka P. Kan-gaslahti of Caltech and Richard Lai and Xi-aobing Mei of Northrop Grumman Corpora-tion for NASA’s Jet Propulsion Laboratory.For more information, contact [email protected]. NPO-47443

180-GHz I-Q Second Harmonic Resistive Mixer MMICNASA’s Jet Propulsion Laboratory, Pasadena, California


used to generate higher frequency signalsin nonlinear Schottky diode-based LOsources. By advancing PA technology, theLO system performance can be increasedwith possible cost reductions compared tocurrent GaAs PAs.

High-power, high-efficiency GaN PAsare cross-cutting and can enable more

efficient local oscillator distributionsystems for new astrophysics and plane-tary receivers and heterodyne array in-struments. It can also allow for a new,electronically scannable solid-statearray technology for future Earth sci-ence radar instruments and communi-cations platforms.

This work was done by King Man Fung,Lorene A. Samoska, Pekka P. Kangaslahti, BjornH. Lambrigtsen, Paul F. Goldsmith, Robert H.Lin, Mary M. Soria, and Joelle T. Cooperrider ofCaltech and Miroslav Micovic and Ara Kur-doghlian of HRL Laboratories for NASA’s JetPropulsion Laboratory. Further information iscontained in a TSP (see page 1). NPO-47364


Electronics/Computers

Compact amplifier assemblies of a typenow being developed for operation at fre-quencies of hundreds of gigahertz com-prise multiple amplifier units in parallelarrangements to increase power and/orcascade arrangements to increase gains.Each amplifier unit is a monolithic mi-crowave integrated circuit (MMIC) imple-mentation of a pair of amplifiers in differ-ential (in contradistinction tosingle-ended) configuration.

Heretofore, in cascading amplifiers toincrease gain, it has been common prac-tice to interconnect the amplifiers by useof wires and/or thin films on substrates.This practice has not yielded satisfactoryresults at frequencies >200 Hz, in eachcase, for either or both of two reasons: • Wire bonds introduce large disconti-

nuities. • Because the interconnections are typi-

cally tens of wavelengths long, any im-pedance mismatches give rise to rip-ples in the gain-vs.-frequency response,which degrade the performance of thecascade.Heretofore, it has been very difficult

to achieve net increases in power bycombining the outputs of amplifiers atfrequencies >100 GHz. The only success-

ful approach that has been even margin-ally successful has involved the use ofwaveguide combiners designed and fab-ricated as components separate from theamplifiers. At these frequencies, evenwaveguides exhibit high losses that caneasily dissipate any power gained bycombining outputs.

In the present development, neitherthin-film nor wire interconnections areused for cascading, and separate compo-nent waveguide combiners are not usedfor combining power. Instead, the ampli-fier units are designed integrally with thewaveguides and designed to be embeddedin the waveguides. The underlying con-cept of differential-amplifier MMICs de-signed integrally with and embedded inwaveguides and the advantages of the dif-ferential over the single-ended configura-tion were reported in “Differential InPHEMT MMIC Amplifiers Embedded inWaveguides” (NPO-42857) NASA TechBriefs, Vol. 33, No. 9 (September 2009),page 35. The novel aspect of the presentdevelopment lies in combining the inte-gration and embedment concepts with thecascading and parallel-combining con-cepts to obtain superior performance. Inan amplifier assembly of the present type,

there is no need for interconnecting wiresor thin-film conductors nor for separatewaveguide power combiners because thewaveguide in which the MMICs are em-bedded is the connecting medium. Be-cause the distance between successiveMMICs in a cascade is only a fraction of awavelength, cascading can be highly effi-cient and ripple in gain versus frequency isreduced to a minimum. Moreover, powercombining can be highly efficient becauseit is accomplished simply by placingMMICs side by side within the waveguide.

This work was done by Pekka Kangaslahtiand Erich Schlecht of Caltech for NASA’s JetPropulsion Laboratory. For more information,contact [email protected].

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 91109-8099E-mail: [email protected] to NPO-44394, volume and number

of this NASA Tech Briefs issue, and thepage number.

Multiple Differential-Amplifier MMICs Embedded in Waveguides Amplifiers are separated by no more than fractional-wavelength distances. NASA’s Jet Propulsion Laboratory, Pasadena, California

Rapid Corner Detection Using FPGAs This algorithm can be used in automotive, navigation, and industrial factory applications. NASA’s Jet Propulsion Laboratory, Pasadena, California

In order to perform precision land-ings for space missions, a control systemmust be accurate to within ten meters.Feature detection applied against im-ages taken during descent and corre-lated against the provided base image iscomputationally expensive and requirestens of seconds of processing time to dojust one image while the goal is toprocess multiple images per second.

To solve this problem, this algo-rithm takes that processing load fromthe central processing unit (CPU) andgives it to a reconfigurable field pro-

grammable gate array (FPGA), whichis able to compute data in parallel atvery high clock speeds. The workloadof the processor then becomes sim-pler; to read an image from a camera,it is transferred into the FPGA, andthe results are read back from theFPGA.

The Harris Corner Detector uses thedeterminant and trace to find a “cor-ner score,” with each step of the com-putation occurring on independentclock cycles. Essentially, the image isconverted into an x and y derivative

map. Once three lines of pixel infor-mation have been queued up, validpixel derivatives are clocked into theproduct and averaging phase of thepipeline. Each x and y derivative issquared against itself, as well as theproduct of the ix and iy derivative, andeach value is stored in a W×N sizebuffer, where W represents the size ofthe integration window and N is thewidth of the image. In this particularcase, a window size of 5 was chosen,and the image is 640×480.

Over a W×N size window, an equidis-


tance Gaussian is applied (to bring outthe stronger corners), and then eachvalue in the entire window is summedand stored. The required componentsof the equation are in place, and it isjust a matter of taking the determinantand trace. It should be noted that thetrace is being weighted by a constant κ,a value that is found empirically to be

within 0.04 to 0.15 (and in this imple-mentation is 0.05). The constant κ de-termines the number of corners avail-able to be compared against athreshold σ to mark a “valid corner.”

After a fixed delay from when the firstpixel is clocked in (to fill the pipeline),a score is achieved after each successiveclock. This score corresponds with an

(x,y) location within the image. If thescore is higher than the predeterminedthreshold σ, then a flag is set high andthe location is recorded.

This work was done by Arin C. Morfopou-los and Brandon C. Metz of Caltech forNASA’s Jet Propulsion Laboratory. Furtherinformation is contained in a TSP (see page1). NPO-47202

Special Component Designs for Differential-Amplifier MMICs Transistors and transmission lines are optimized for differential operation. NASA’s Jet Propulsion Laboratory, Pasadena, California

Special designs of two types of elec-tronic components — transistors andtransmission lines — have been con-ceived to optimize the performances ofthese components as parts of waveguide-embedded differential-amplifier mono-lithic microwave integrated circuits(MMICs) of the type described in theimmediately preceding article. Thesedesigns address the following two issues,the combination of which is unique tothese particular MMICs: • Each MMIC includes a differential

double-strip transmission line that typ-ically has an impedance between 60and 100 W. However, for purposes ofmatching of impedances, transmissionlines having lower impedances arealso needed.

• The transistors in each MMIC are,more specifically, one or more pair(s)of InP-based high-electron-mobilitytransistors (HEMTs). Heretofore, it hasbeen common practice to fabricate eachsuch pair as a single device configuredin the side-to-side electrode sequencesource/gate/drain/gate/source. Thisconfiguration enables low-inductancesource grounding from the sides of

the device. However, this configura-tion is not suitable for differential op-eration, in which it is necessary todrive the gates differentially and tofeed the output from the drain elec-trodes differentially.The special transmission-line design

provides for three conductors, insteadof two, in places where lower imped-ance is needed. The third conductor isa metal strip placed underneath the dif-ferential double-strip transmission line.The third conductor increases the ca-pacitance per unit length of the trans-mission line by such an amount as to re-duce the impedance to between 5 and15 W.

In the special HEMT-pair design, theside-to-side electrode sequence ischanged to drain/gate/source/gate/drain. In addition, the size of the sourceis reduced significantly, relative to corre-sponding sizes in prior designs. This re-duction is justified by the fact that, byvirtue of the differential configuration,the device has an internal virtualground, and therefore there is no needfor a low-resistance contact between thesource and the radio-frequency circuitry.

The source contact is needed only forDC biasing.

These designs were implemented in asingle-stage-amplifier MMIC. In a test ata frequency of 305 GHz, the amplifierembedded in a waveguide exhibited again of 0 dB; after correcting for the lossin the waveguide, the amplifier wasfound to afford a gain of 0.9 dB. In a testat 220 GHz, the overall gain of the am-plifier-and-waveguide assembly wasfound to be 3.5 dB.

This work was done by Pekka Kangaslahtiof Caltech for NASA’s Jet Propulsion Labora-tory. For more information, [email protected].




A multi-stage automated target recog-nition (ATR) system has been designedto perform computer vision tasks with ad-equate proficiency in mimicking humanvision. The system is able to detect, iden-tify, and track targets of interest. Potentialregions of interest (ROIs) are first identi-

fied by the detection stage using an Opti-mum Trade-off Maximum Average Corre-lation Height (OT-MACH) filter com-bined with a wavelet transform. Falsepositives are then eliminated by the veri-fication stage using feature extractionmethods in conjunction with neural net-

works. Feature extraction transforms theROIs using filtering and binning algo-rithms to create feature vectors. A feed-forward back-propagation neural net-work (NN) is then trained to classify eachfeature vector and to remove false posi-tives. The system parameter optimiza-

Multi-Stage System for Automatic Target RecognitionThis system is capable of identifying hazards to avoid in robotic vehicle and automobilenavigation.NASA’s Jet Propulsion Laboratory, Pasadena, California

tions process has been developed toadapt to various targets and datasets.

The objective was to design an efficientcomputer vision system that can learn todetect multiple targets in large imageswith unknown backgrounds. Because thetarget size is small relative to the imagesize in this problem, there are many re-gions of the image that could potentiallycontain the target. A cursory analysis ofevery region can be computationally effi-cient, but may yield too many false posi-tives. On the other hand, a detailed analy-sis of every region can yield better results,but may be computationally inefficient.The multi-stage ATR system was designedto achieve an optimal balance between

accuracy and computational efficiency byincorporating both models.

The detection stage first identifies po-tential ROIs where the target may bepresent by performing a fast Fourier do-main OT-MACH filter-based correlation.Because threshold for this stage is cho-sen with the goal of detecting all truepositives, a number of false positives arealso detected as ROIs. The verificationstage then transforms the regions of in-terest into feature space, and eliminatesfalse positives using an artificial neuralnetwork classifier.

The multi-stage system allows tuningthe detection sensitivity and the identifi-cation specificity individually in each

stage. It is easier to achieve optimizedATR operation based on its specific goal.The test results show that the system wassuccessful in substantially reducing thefalse positive rate when tested on a sonarand video image datasets.

This work was done by Tien-Hsin Chao,Thomas T. Lu, and David Ye of Caltech; We-ston Edens of Butler University; and OliverJohnson of Harvey Mudd College for NASA’sJet Propulsion Laboratory. For more informa-tion, contact [email protected].

The software used in this innovation isavailable for commercial licensing. Please con-tact Daniel Broderick of the California Insti-tute of Technology at [email protected] to NPO-47012.

The Multi-Stage ATR System Architecture incorporates a detection stage that first identifies potential ROIs where the target may be present by perform-ing a Fast Fourier domain OT-MACH filter-based correlation.

Training Images

Filter Synthesizer

OT-MACHFilter Bank

Correlator

Ground Truth

Match

Next Image

No

YesFeature Extraction

Neural NetworkTraining & ID

ROIs

Detected ROI Coordinates

2nd Stage Fine Target ID

ID Results

1st Stage Course Target Detection


Existing software has been modifiedto yield the benefits of integer fixeddouble-differenced GPS-phased ambi-guities when processing data from asingle GPS receiver with no access toany other GPS receiver data. When thedouble-differenced combination ofphase biases can be fixed reliably, a sig-

nificant improvement in solution accu-racy is obtained.

This innovation uses a large global setof GPS receivers (40 to 80 receivers) tosolve for the GPS satellite orbits andclocks (along with any other parame-ters). In this process, integer ambiguitiesare fixed and information on the ambi-

guity constraints is saved. For each GPStransmitter/receiver pair, the processsaves the arc start and stop times, thewide-lane average value for the arc, thestandard deviation of the wide lane, andthe dual-frequency phase bias after biasfixing for the arc. The second step of theprocess uses the orbit and clock infor-

Single-Receiver GPS Phase Bias Resolution NASA’s Jet Propulsion Laboratory, Pasadena, California


mation, the bias information from theglobal solution, and only data from thesingle receiver to resolve double-differ-enced phase combinations. It is called“resolved” instead of “fixed” becauseconstraints are introduced into theproblem with a finite data weight to bet-ter account for possible errors.

A receiver in orbit has much shortercontinuous passes of data than a receiverfixed to the Earth. The method has pa-

rameters to account for this. In particu-lar, differences in drifting wide-lane val-ues must be handled differently. Thefirst step of the process is automated,using two JPL software sets, Longarc andGipsy-Oasis. The resulting orbit/clockand bias information files are posted onanonymous ftp for use by any licensedGipsy-Oasis user. The second step is im-plemented in the Gipsy-Oasis exe-cutable, gd2p.pl, which automates the

entire process, including fetching the in-formation from anonymous ftp.

This work was done by William I Bertiger,Bruce J. Haines, Jan P. Weiss, and NathanielE. Harvey of Caltech for NASA’s Jet Propul-sion Laboratory. For more information, con-tact [email protected].

This software is available for commercial li-censing. Please contact Daniel Broderick ofthe California Institute of Technology [email protected]. Refer to NPO-47149.

Systems that measure the angles of ar-rival of ultra-wideband (UWB) radio sig-nals and perform triangulation by use ofthose angles in order to locate the sourcesof those signals are undergoing develop-ment. These systems were originally in-tended for use in tracking UWB-transmit-ter-equipped astronauts and mobilerobots on the surfaces of remote planetsduring early stages of exploration, beforesatellite-based navigation systems becomeoperational. On Earth, these systemscould be adapted to such uses as trackingUWB-transmitter-equipped firefighters in-side buildings or in outdoor wildfire areasobscured by smoke.

The same characteristics that havemade UWB radio advantageous for fine-resolution ranging, covert communica-tion, and ground-penetrating radar ap-plications in military andlaw-enforcement settings also contributeto its attractiveness for the present track-ing applications. In particular, the wave-form shape and the short duration ofUWB pulses make it possible to attainthe high temporal resolution (of theorder of picoseconds) needed to meas-ure angles of arrival with sufficient preci-sion, and the low power spectral densityof UWB pulses enables UWB radio com-munication systems to operate in prox-imity to other radio communication sys-tems with little or no perceptible mutualinterference.

The figure schematically depicts asimple system of this type engaged intracking a single UWB transmitter on aplane. The system includes two UWB-re-ceiver assemblies, denoted clusters, sep-arated by a known length d. Within eachcluster is a UWB receiver connected totwo antennas that are separated by alength a that is much shorter than the

aforementioned length d. The signalsreceived by the two antennas in eachcluster are subjected to a process ofcross-correlation plus peak detection tomeasure differences between theirtimes of arrival. It is assumed that thedistances (r1 and r2) between the clus-ters and the transmitter are muchgreater than a, as would usually be thecase in most practical applications.Then the angles of arrival of the signalsat the clusters are given by θ1≈arccos(cτ1/a) and θ2≈arccos (cτ2/a); where θ1and θ2 are as shown in the figure; c is the

speed of light; τ1 is the difference be-tween the times of arrival of a pulse atthe antennas in cluster 1; and τ2 is thedifference between the times of arrivalof a pulse at the antennas in cluster 2.Then using θ1 and θ2, the two-dimen-sional location of the transmitter, rela-tive to the known locations of the clus-ters, is calculated straightforwardly byuse of the triangulation equations.

The processing of signals to deter-mine the differences between theirtimes of arrival, and the subsequent pro-cessing to determine the angles of ar-

Ultra-Wideband Angle-of-Arrival Tracking SystemsUWB radio pulses afford temporal resolution needed for estimating angles of arrival.Lyndon B. Johnson Space Center, Houston, Texas

UWBReceiver

Cluster1

Cluster2

UWBTransmitter

θ1

r1

r2

a

UWBReceiver

θ2

a

d

Angles θ1 and θ2 Are Estimated from differences between the times of arrival of UWB radio pulses atthe antennas in each cluster. Then using these angles, the relative position of the transmitter is calcu-lated by triangulation.


There is an update on the subject mat-ter of “Differential InP HEMT MMICAmplifiers Embedded in Waveguides”(NPO-42857) NASA Tech Briefs, Vol. 33,No. 9 (September 2009), page 35. To re-capitulate: Monolithic microwave inte-grated-circuit (MMIC) amplifiers of atype now being developed for operationat frequencies of hundreds of gigahertzcontain InP high-electron-mobility tran-sistors (HEMTs) in a differential config-uration. The MMICs are designed inte-grally with, and embedded in, waveguidepackages. The instant work does notmention InP HEMTs but otherwise reit-erates part of the subject matter of thecited prior article, with emphasis on thefollowing salient points:

• An MMIC is mounted in the elec-tric-field plane (“E-plane”) of awaveguide and includes a finlinetransition to each differential-am-plifier stage.

• The differential configuration createsa virtual ground within each pair oftransistor-gate fingers, eliminating theneed for external radio-frequencygrounding. This work concludes by describing a

single-stage differential submillimeter-wave amplifier packaged in a rectangu-lar waveguide and summarizing resultsof tests of this amplifier at frequencies of220 and 305 GHz.

This work was done by Pekka Kangaslahtiand Erich Schlecht of Caltech for NASA’s Jet

Propulsion Laboratory. For more information,contact [email protected].




Update on Waveguide-Embedded Differential MMIC AmplifiersNASA’s Jet Propulsion Laboratory, Pasadena, California

rival and the position of the transmitter,is done in a computer external to theclusters. For this purpose, the receivedwaveforms are digitized in the receivers,and the waveform data are sent to thecomputer via a hub. Even though no at-tempt is made to synchronize operationof the two receivers, the data from thereceivers are quasi-synchronized bymeans of interface software that effectsparallel socket communication with datasegmentation, summarized as follows:Waveform data are collected from eachreceiver in segments, whenever they be-come available and the computer is

ready to collect them. The segmentsfrom each receiver are labeled as havingcome from that receiver and, in the col-lection process, are interleaved withthose from the other receiver in chrono-logical order of collection. Within thecomputer, the segments from each re-ceiver are stored in a separate buffer.Thus, the contents of the buffers arerepresentations of the same UWB pulsewaveform arriving at the two receivers atapproximately the same time. When thebuffers for both receivers contain com-plete representations of a UWB pulsewaveform, the data from that buffer are

copied into an array for use in the calcu-lations described above.

This work was done by G. Dickey Arndt,Phong H. Ngo, Chau T. Phan, and JuliaGross of Johnson Space Center; Jianjun NiNRC fellow; and John Dusl of Jacobs Sver-drup. Further information is contained in aTSP (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, Johnson SpaceCenter, (281) 483-1003. Refer to MSC-24184-1.


Automation Framework forFlight Dynamics ProductsGeneration

XFDS provides an easily adaptable au-tomation platform. To date it has beenused to support flight dynamics opera-tions. It coordinates the execution ofother applications such as SatelliteTookKit, FreeFlyer, MATLAB, and Perlcode. It provides a mechanism for pass-ing messages among a collection ofXFDS processes, and allows sending andreceiving of GMSEC messages. A unifiedand consistent graphical user interface(GUI) is used for the various tools. Itsautomation configuration is stored intext files, and can be edited either di-rectly or using the GUI.

XFDS is implemented as a group ofcooperating processes. One process co-ordinates communications, anotherdrives an optional GUI (not needed ifrunning in batch mode), and the restcarry out automation tasks. The soft-ware is designed around three con-cepts: (1) an “action” controls an au-tomation step; (2) a “variable” allowsinformation to be shared among ac-tions; and (3) a “form” corresponds to aGUI widget, which can be reused be-tween action editors.

A significant strength of this approachis to provide a high-level abstraction tothe procedures that need to be carriedout. Frequently changed parameters arereadily available for modification, whilethe rest are hidden. Additional pro-grams that provide a batch interface canbe added to this system.

This work was done by Robert E. Wiegand,Timothy C. Esposito, John S. Watson, LindaJun, Wendy Shoan, and Carla Matusow ofGoddard Space Flight Center and Wayne Mc-Cullough of Computer Sciences Corp. For fur-ther information, contact the Goddard Inno-vative Partnerships Office at (301)286-5810. GSC-15618-1.

Product Operations StatusSummary Metrics

The Product Operations Status Sum-mary Metrics (POSSUM) computer pro-gram provides a readable view into thestate of the Phoenix Operations ProductGeneration Subsystem (OPGS) datapipeline. POSSUM provides a user inter-face that can search the data store, collect

product metadata, and display the resultsin an easily-readable layout. It was de-signed with flexibility in mind for supportin future missions. Flexibility over variousdata store hierarchies is providedthrough the disk-searching facilities ofMarsviewer. This is a proven programthat has been in operational use since thefirst day of the Phoenix mission.

POSSUM presents a graphical repre-sentation for tracking and accountabilityof the existence and version of expectedReduced Data Records (RDRs) for agiven Experiment Data Record (EDR)for the Phoenix lander mission. By usingPOSSUM, operations personnel can eas-ily determine if any RDRs are missing fora given EDR during and after OPGS pro-cessing pipeline. A variety of sort op-tions exists, including product file cre-ation time and instrument.

This work was done by Atsuya Takagi andNicholas Toole of Caltech for NASA’s JetPropulsion Laboratory.


Mars Terrain GenerationA suite of programs for the generation

of disparity maps from stereo imagepairs via correlation, and conversion ofthose disparity maps to XYZ maps, hasbeen updated. This suite implements anautomated method of deriving terrainfrom stereo images for use in theground data system for in-situ (landerand rover) cameras. This differs fromonboard correlation by concentratingmore on accuracy than speed, sincenear-real-time is not a requirement onthe ground. The final result is an XYZvalue for every point in the image thatpasses several quality checks. A priorigeometric camera calibration informa-tion is required for this suite to operate.

The suite is very flexible, enabling itsuse in many special situations, such asnon-linearized images required for ap-plications like the Phoenix arm camera,or long-baseline stereo, where the rovermoves between left and right images.The suite consists of: • marscor3: The primary correlation

program used by MER, PHX, and(soon) MSL. Requires a low-resolution

disparity map as input, and refines it. • marsjplstereo: A wrapper around a

much faster correlator that assumesthe images are epipolar aligned (“lin-earized”). Creates the input formarscor3.

• marsunlinearize: Takes a linearizedcorrelation result and unprojects itback to raw geometry. Creates theinput for marscor3 in some non-lin-earized situations.

• marsfakedisp: Creates the input formarscor3 in some non-linearized casesby assuming an approximate surface.

• marsdispinvert: “Inverts” a disparitymap (e.g., from L→R to R→L) to cre-ate an input for marscor3.

• marsxyz: Takes the disparity map (e.g.,from marscor3) and generates XYZ co-ordinates for each pixel.

• marsfilter: Filters the output ofmarsxyz to mask off undesirable areas(such as the rover itself, horizon, etc.).

• marsrange: Takes the results frommarsxyz and computes range from thecamera for each pixel. While the underlying correlation coef-

ficient computation is nearly the same asin the original, the algorithms drivinghow that correlation is accomplishedhave been completely redesigned. In ad-dition, significant new capability exists,such as non-linearized stereo, R→L in-version, masking, and range computa-tion. Significant additions to marsxyz’sfiltering capability to reject bad correla-tions have also been made.

This work was done by Robert G. Deen ofCaltech for NASA’s Jet Propulsion Laboratory.

This software is available for commercial li-censing. Please contact Daniel Broderick ofthe California Institute of Technology [email protected]. Refer to NPO-46659.

Application-Controlled Par-allel AsynchronousInput/Output Utility

A software utility tool has been de-signed to alleviate file system I/O per-formance bottlenecks to which manyhigh-end computing (HEC) applica-tions fall prey because of the relativelylarge volume of data generated for agiven amount of computational work. Inan effort to reduce computing resourcewaste, and to improve sustained per-formance of these HEC applications, alightweight software utility has been de-

Software


signed to circumvent bandwidth limita-tions of typical HEC file systems by ex-ploiting the faster inter-processor band-width to move output data fromcompute nodes to designated I/O nodesas quickly as possible, thereby minimiz-ing the I/O wait time. This utility hassuccessfully demonstrated a significantperformance improvement within amajor NASA weather application.

This work was done by Thomas Clune ofGoddard Space Flight Center and ShujiaZhou of Northrop Grumman. For further in-formation, contact the Goddard InnovativePartnerships Office at (301) 286-5810. GSC-15751-1

Planetary Image GeometryLibrary

The Planetary Image Geometry (PIG)library is a multi-mission library used forprojecting images (EDRs, or Ex per i -ment Data Records) and managingtheir geometry for in-situ missions. Acollection of models describes camerasand their articulation, allowing applica-tion programs such as mosaickers, ter-rain generators, and pointing correc-

tion tools to be written in a multi-mis-sion manner, without any knowledge ofparameters specific to the supportedmissions.

Camera model objects allow transfor-mation of image coordinates to andfrom view vectors in XYZ space. Point-ing models, specific to each mission, de-scribe how to orient the camera modelsbased on telemetry or other informa-tion. Surface models describe the sur-face in general terms. Coordinate sys-tem objects manage the variouscoordinate systems involved in mostmissions. File objects manage access tometadata (labels, including telemetryinformation) in the input EDRs andRDRs (Reduced Data Records). Labelmodels manage metadata informationin output files. Site objects keep track ofdifferent locations where the spacecraftmight be at a given time. Radiometrymodels allow correction of radiometryfor an image. Mission objects containbasic mission parameters. Pointing ad-justment (“nav”) files allow pointing tobe corrected.

The object-oriented structure (C++)makes it easy to subclass just the pieces

of the library that are truly mission-spe-cific. Typically, this involves just thepointing model and coordinate systems,and parts of the file model. Once the li-brary was developed (initially for MarsPolar Lander, MPL), adding new mis-sions ranged from two days to a fewmonths, resulting in significant cost sav-ings as compared to rewriting all the ap-plication programs for each mission.Currently supported missions includeMars Pathfinder (MPF), MPL, Mars Ex-ploration Rover (MER), Phoenix, andMars Science Lab (MSL). Applicationsbased on this library create the majorityof operational image RDRs for thosemissions. A Java wrapper around the li-brary allows parts of it to be used fromJava code (via a native JNI interface). Fu-ture conversions of all or part of the li-brary to Java are contemplated.

This work was done by Robert G. Deen andOleg Pariser of Caltech for NASA’s Jet Propul-sion Laboratory.



Manufacturing & Prototyping

The nation is challenged to decreasethe cost and schedule to develop newspace transportation propulsion systemsfor commercial, scientific, and militarypurposes. Better design criteria andmanufacturing techniques for smallthrusters are needed to meet currentapplications in missile defense, space,and satellite propulsion. The require-ments of these systems present size, per-formance, and environmental demandson these thrusters that have posed sig-nificant challenges to the current de-signers and manufacturers. Designersare limited by manufacturing processes,which are complex, costly, and timeconsuming, and ultimately limited intheir capabilities.

The PDFF innovation vastly extendsthe design opportunities of rocket en-gine components and systems by mak-ing use of the unique manufacturingfreedom of solid freeform rapid proto-type manufacturing technology com-bined with the benefits of ceramic ma-terials. The unique features of PDFFare developing and implementing a de-sign methodology that uses solidfreeform fabrication (SFF) techniquesto make propulsion components with

significantly improved performance,thermal management, power density,and stability, while reducing develop-ment and production costs. PDFF ex-tends the design process envelope be-yond conventional constraints byleveraging the key feature of the SFFtechnique with the capability to formobjects with nearly any geometric com-plexity without the need for elaboratemachine setup. The marriage of SFFtechnology to propulsion componentsallows an evolution of design practiceto harmonize material properties withfunctional design efficiency.

Reduced density of materials whencoupled with the capability to honey-comb structure used in the injector willhave significant impact on overall massreduction. Typical thrusters in use forattitude control have 60–90 percent ofits mass in the valve and injector, whichis typically made from titanium. Thecombination of material and structureenvisioned for use in an SFF thrusterdesign could reduce thruster weight bya factor of two or more. The thrust-to-weight ratios for such designs canachieve 1,000:1 or more, depending onchamber pressure.

The potential exists for continued de-velopment in materials, size, speed, ac-curacy of SFF techniques, which canlead to speculative developments ofPDFF processes such as fabrication ofcustom human interface devices likemasks, chairs, and clothing, and ad-vanced biomedical application tohuman organ reconstruction.

Other potential applications are:higher fidelity lower cost test fixtures forprobes and inspection, disposablethrusters, and ISRU (in situ resource uti-lization) for component production inspace or on Lunar and Martian missions,and application for embedding MEMS(microelectromechanical systems) dur-ing construction process of form chang-ing aerostructure/dynamic structures.

This work was done by Daudi Barnes ofDMX Engineering, Jim McKinnon of FrontierEngineering, and Richard Priem of Priem Con-sultants for Glenn Research Center. Further in-formation is contained 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: Steven Fedor, MailStop 4–8, 21000 Brookpark Road, Cleve-land, Ohio 44135. Refer to LEW-18557-1.

Propulsion Design With Freeform Fabrication (PDFF)Innovation for ceramic materials uses solid freeform rapid prototype manufacturing technology. John H. Glenn Research Center, Cleveland, Ohio

A method was developed for produc-ing thick-section [>2 in. (≈5 cm)], con-tinuous fiber-reinforced ceramic matrixcomposites (CMCs). Ultramet-modifiedfiber interface coating and melt infiltra-tion processing, developed previouslyfor thin-section components, were usedfor the fabrication of CMCs that were anorder of magnitude greater in thickness[up to 2.5 in. (≈6.4 cm)]. Melt process-

ing first involves infiltration of a fiberpreform with the desired interface coat-ing, and then with carbon to partiallydensify the preform. A molten refractorymetal is then infiltrated and reacts withthe excess carbon to form the carbidematrix without damaging the fiber rein-forcement. Infiltration occurs from theinside out as the molten metal fills virtu-ally all the available void space. Densifi-

cation to <5 vol% porosity is a one-stepprocess requiring no intermediate ma-chining steps.

The melt infiltration method requiresno external pressure. This prevents over-infiltration of the outer surface plies,which can lead to excessive residualporosity in the center of the part. How-ever, processing of thick-section compo-nents required modification of the con-

Economical Fabrication of Thick-Section Ceramic Matrix Composites Applications for these composites include combustors, high-temperature filter elements, andprocess industry parts requiring corrosion resistance. Marshall Space Flight Center, Alabama


ventional process conditions, and themeans by which the large amount ofmolten metal is introduced into the fiberpreform. Modification of the low-tem-perature, ultraviolet-en hanced chemicalvapor deposition process used to applyinterface coatings to the fiber preformwas also required to accommodate thehigh preform thickness.

The thick-section CMC processing de-veloped in this work proved to be in-valuable for component development,fabrication, and testing in two comple-

mentary efforts. In a project for theArmy, involving SiC/SiC blisk develop-ment, nominally 0.8 in. thick × 8 in. di-ameter (≈2 cm thick × 20 cm diameter)components were successfully infil-trated. Blisk hubs were machined usingdiamond-embedded cutting tools andsuccessfully spin-tested. Good ply uni-formity and extremely low residualporosity (<2 percent) were achieved,the latter being far lower than thatachieved with SiC matrix compositesfabricated via CVI or PIP. The pyrolytic

carbon/zirconium nitride interfacecoating optimized in this work for useon carbon fibers was incorporated inthe SiC/SiC composites and yielded a>41 ksi (≈283 MPa) flexural strength.

This work was done by Jason Babcock,Gautham Ramachandran, Brian Williams,and Robert Benander of Ultramet for Mar-shall Space Flight Center. For more informa-tion, contact Sammy Nabors, MSFC Commer-cialization Assistance Lead, [email protected]. Refer to MFS-32736-1.

To produce a noble metal-on-metaloxide catalyst on an inert, high-surface-area support material (that functions asa catalyst at approximately room temper-ature using chloride-free rea gents), foruse in a carbon dioxide laser, requirestwo steps: First, a commercially available,inert, high-surface-area support material(silica spheres) is coated with a thinlayer of metal oxide, a monolayer equiv-alent. Very beneficial results have beenobtained using nitric acid as an oxidiz-ing agent because it leaves no residue. Itis also helpful if the spheres are first de-aerated by boiling in water to allow theentire surface to be coated. A metal,such as tin, is then dissolved in the oxi-dizing agent/support material mixtureto yield, in the case of tin, metastannicacid. Although tin has proven especiallybeneficial for use in a closed-cycle CO2laser, in general any metal with two va-lence states, such as most transition met-als and antimony, may be used. Themetastannic acid will be adsorbed ontothe high-surface-area spheres, coatingthem. Any excess oxidizing agent is thenevaporated, and the resulting metastan-

nic acid-coated spheres are dried andcalcined, whereby the metastannic acidbecomes tin(IV) oxide.

The second step is accomplished bypreparing an aqueous mixture of thetin(IV) oxide-coated spheres, and a solu-ble, chloride-free salt of at least one cat-alyst metal. The catalyst metal may be se-lected from the group consisting ofplatinum, palladium, ruthenium, gold,and rhodium, or other platinum groupmetals. Extremely beneficial results havebeen obtained using chloride-free saltsof platinum, palladium, or a combina-tion thereof, such as tetraammineplat-inum (II) hydroxide ([Pt(NH3)4](OH)2), or tetraammine palladium ni-trate ([Pd(NH3)4](NO3)2).

It is also beneficial if the coatedspheres are first de-aerated by boiling.The platinum salt will be adsorbed ontothe coated spheres. A chloride-free re-ducing agent is then added to the aque-ous mixture whereby the catalyst metal isdeposited on the tin(IV) oxide-coatedspheres. Any reducing agent that de-composes to volatile products and waterupon reaction or drying may be used.

Formic acid, hydroxylamine (NH2OH),hyrdrazine (N2H4), and ascorbic acidare particularly advantageous. After themetal has been deposited on the tin(IV)oxide-coated spheres, the solution isevaporated to dryness, whereby the de-sired noble metal-on-metal oxide cata-lyst is obtained.

This innovative process results in amore uniform application than othermethods. Similarly, the method of form-ing and applying a precious metal to ei-ther tin oxide, or an inert substrate, is aone-step process and occurs at a lowertemperature than that commonly usedby other processes. This invention is in-herently clean because excess reagents,such as nitric acid and formic acid, aswell as unwanted products, such as ni-trates and formates, all decompose andare removed from the system by simpleevaporation without the necessity forseparating them by filtration or washing.

This work was done by Patricia Davis andIrvin Miller of Langley Research Center andBilly Upchurch of Science and TechnologyCorporation. Further information is con-tained in a TSP (see page 1). LAR-13741-1

Process for Making a Noble Metal on Tin Oxide Catalyst This method produces an efficient, room-temperature catalyst for recombining carbonmonoxide and oxygen products. Langley Research Center, Hampton, Virginia


Mechanics/Machinery

Some refinements have been conceivedfor a proposed apparatus that would gen-erate hydrogen (for use in a fuel cell) bymeans of chemical reactions among mag-nesium, magnesium hydride, and steam.The refinements lie in tailoring spatialand temporal distributions of steam andliquid water so as to obtain greater overallenergy-storage or energy-generation effi-ciency than would otherwise be possible.

A description of the prior art is pre-requisite to a meaningful description ofthe present refinements. The hydrogen-generating apparatus in question is oneof two versions of what was called the“advanced hydrogen generator” in“Fuel-Cell Power Systems IncorporatingMg-Based H2 Generators” (NPO-43554),NASA Tech Briefs, Vol. 33, No. 1 (January2009), page 52. To recapitulate: The ap-paratus would include a reactor vesselthat would be initially charged with mag-nesium hydride. The apparatus wouldexploit two reactions: • The endothermic decomposition reac-

tion MgH2 → Mg + H2, which occurs ata temperature ≥300 °C, and

• The exothermic oxidation reactionMgH2 + H2O → MgO + 2H2, which oc-curs at a temperature ≥330 °C.Once the initial heating was com-

plete and both reactions under way, theendothermic reaction would be sus-tained by the heat generated from theexothermic reaction. For every mole ofMgH2 oxidized, sufficient waste heat is

generated to decompose an additionalthree moles of the hydride. As a conse-quence of these reaction ratios, themajor reaction product is Mg, and theminor one MgO. Both have extremelylow toxicity. MgH2 is easily recycled toMg. In theory, no energy is required be-cause regeneration produces enough

heat to power the process. A practicalsystem would not be 100-percent effi-cient so it would be expected that therewould be a modest energy cost. TheMgO can be safely and easily recycledin a magnesium-refining plant for lessthan the cost of smelting Mg becauseMgO is an intermediate product of that

Refinements in an Mg/MgH2/H2O-Based Hydrogen Generator Externally generated steam would be needed only briefly to start operation. NASA’s Jet Propulsion Laboratory, Pasadena, California

Water Would Be Injected at one end of a reactor bed containing MgH2 particles. Initially, the waterwould be in the form of steam as needed to start the reactions. Thereafter, liquid water would be in-jected, and one of the reactions, which is exothermic, would supply the necessary heat.

MgO

MgH2MgH2 MgH2MgH2

Reaction Zone

SteamFront

SteamorLiquid Water

BEFORE OPERATION DURING OPERATION

This Brief describes a method of ma-chining and assembly when the depth ofcorrugations far exceeds the width andconventional machining is not practical.The horn is divided into easily machined,individual rings with shoulders to controlthe depth. In this specific instance, eachof the corrugations is identical in profile,and only differs in diameter and outerprofile. The horn is segmented into rings

that are cut with an interference fit (zeroclearance with all machining errors bi-ased toward contact). The interferencefaces can be cut with a reverse taper to in-crease the holding strength of the joint.The taper is a compromise between theinterference fit and the clearance of thetwo faces during assembly.

Each internal ring is dipped in liquidnitrogen, then nested in the previous,

larger ring. The ring is rotated in the nestuntil the temperature of the two partsequalizes and the pieces lock together.The resulting assay is stable, strong, andhas an internal finish that cannot beachieved through other methods.

This work was done by John B. Sosnowskiof Caltech for NASA’s Jet Propulsion Labora-tory. For more information, [email protected]. NPO-47213

Stacked Corrugated Horn Rings NASA’s Jet Propulsion Laboratory, Pasadena, California


process. This concludes the descriptionof the prior art.

The present refinements reflect thefollowing ideas: In principle, the designand operation of the reactor could beoptimized in the sense that the rate ofgeneration of heat in the exothermic re-action, the rate of consumption of heatin the endothermic reaction, and theflow of heat from the exothermic to theendothermic reaction could be tailoredto minimize the input energy needed toproduce a given amount of H2. In turn,the reaction rates could be tailored bytailoring gradients of temperature andchemical composition, which, in turn,could be tailored through adjustmentsin rates of flow of steam and liquid waterinto the reactor, recognizing a need toadjust the rates because the gradients of

temperature and chemical compositionevolve as MH2 is consumed and MgO isgenerated.

For the purpose of illustrating the re-finements, the figure schematically de-picts a reactor that has a simple shapeand inlets for steam and liquid water atone end only. One of the refinementswould be to inject steam only at the be-ginning of operation, in no more thanthe quantity needed to initiate theexothermic reaction. The combinationof heating and water vapor provided bysteam would initiate both exothermicand endothermic reactions. With initia-tion of the exothermic reaction, suffi-cient heat is produced to allow the useof liquid water feed instead of steam.Some of the heat from the exothermicreaction would be consumed in heating

the liquid water to steam. As operationcontinued, the steam front and the hy-drogen-generating region would movefarther into the region initially contain-ing MgH2, leaving behind the MgO andMg waste product. The rate of theexothermic reaction would be adjustedby adjusting the rate of injection of liq-uid water. If the reactor were of a more-complex configuration featuring multi-ple injection points, then the rates ofinjection of water at those points couldbe adjusted individually to obtain amore nearly optimum spatial and tem-poral distribution of temperaturethroughout the reactor.

This work was done by Andrew Kindlerand Yuhong Huang of Caltech for NASA’s JetPropulsion Laboratory. For more information,contact [email protected]. NPO-46064

Continuous/Batch Mg/MgH2/H2O-Based Hydrogen GeneratorSize and weight would be less than those of prior H2 generators.NASA’s Jet Propulsion Laboratory, Pasadena, California

A proposed apparatus for generatinghydrogen by means of chemical reac-tions of magnesium and magnesiumhydride with steam would exploit thesame basic principles as those dis-cussed in the immediately precedingarticle, but would be designed to im-plement a hybrid continuous/batchmode of operation. The design con-cept would simplify the problem of op-timizing thermal management andwould help to minimize the size andweight necessary for generating a givenamount of hydrogen.

The apparatus would include a vessel,the interior volume of which would bedivided into an upper and a lower tank(see figure). The upper tank wouldserve as a fuel-storage/feeder unit: Itwould include a bellows initially filledwith MgH2 powder (the fuel), plus amechanism that would include a rotat-ing threaded outer pipe with meteringwindows and a non-rotating, non-threaded inner pipe with metering win-dows, for feeding the powder into thelower tank at a controlled rate. As theouter pipe was rotated, the widows in thepipes would alternately expose or oc-clude each other. The mechanism wouldbe driven by an external motor via amagnetic coupling. The mechanismwould also serve partly as a valve to pre-vent the undesired flow of steam fromthe reactor into the storage volume in This Hydrogen Generator would function in a hybrid of batch and continuous modes.

Magnetic Coupling

Inlet for MgH2 Powder

MgH2 Powder

Threaded, RotatingOuter Pipe

Manifold forDistributing

Water or Steam

Outlet forWaste Product

Inlet forWater or

Steam

Outlet forHydrogen

Lower Tank

Upper Tank

Bellows

MeteringWindows


that when powder was not being fed, theouter pipe would be rotated to a shaftangle at which the windows in the twopipes would occlude each other. Thethread on the outer pipe would engagea threaded fitting on the bottom of thebellows, so that rotation of the outerpipe would compress the bellows as thepowder was consumed.

The lower tank would serve as both areactor and chamber for storing thesolid waste end product (MgO) of thehydrogen-generating reactions. As thepowder was fed from the upper tank tothe lower tank and the bellows was com-pressed, the volume of the lower tankwould grow, making room for the grow-ing amount of waste material. Becausefresh fuel would be dropped over themost recently reacted portion of theconsumed fuel, it would always come in

contact with the hottest part. Therewould be ample time for the fuel toreact as nearly completely as possiblebecause once the fuel was in the reac-tor, it would stay there. A thermally in-sulating layer (not shown in the figure)on the bottom of the bellows would re-duce the undesired flow of heat fromthe reactor to the storage volume,thereby helping to suppress undesireddecomposition of the MgH2 in the stor-age volume.

As described thus far, the apparatuswould operate in a batch mode. Theupper tank could be refilled with MgH2powder from the top, and the MgOsolid waste could be removed from thebottom. However, it would be necessaryto interrupt operation during such re -filling and emptying and during con-comitant reverse rotation of the

threaded outer pipe to reset the bellowsto full storage volume. Thus, truly con-tinuous operation would not be possi-ble: The apparatus would operate in aquasi-batch, quasi-continuous mode.

This work was done by Andrew Kindlerand Yuhong Huang of Caltech for NASA’s JetPropulsion Laboratory.




The development of the Ko displace-ment theory for predictions of structuredeformed shapes was motivated in 2003 bythe Helios flying wing, which had a 247-ft(75-m) wing span with wingtip deflectionsreaching 40 ft (12 m). The Helios flyingwing failed in midair in June 2003, creat-ing the need to develop new technology topredict in-flight deformed shapes of un-manned aircraft wings for visual displaybefore the ground-based pilots.

Any types of strain sensors installed ona structure can only sense the surface

strains, but are incapable to sense theoverall deformed shapes of structures.After the invention of the Ko displace-ment theory, predictions of structure de-formed shapes could be achieved byfeeding the measured surface strainsinto the Ko displacement transfer func-tions for the calculations of out-of-planedeflections and cross sectional rotationsat multiple locations for mapping outoverall deformed shapes of the struc-tures. The new Ko displacement theorycombined with a strain-sensing system

thus created a revolutionary new struc-ture-shape-sensing technology.

The formulation of the Ko displace-ment theory stemmed from the inte-grations of the beam curvature equa-tion (second order differentialequation). The beamlike structure(wing) was first discretized into multi-ple small domains so that beam depthand surface strain distributions couldbe represented with piecewise linearfunctions. This discretization approachenabled piecewise integrations of the

Ko Displacement Theory for Structural Shape PredictionsPrediction system enables real-time aero-elastic aircraft wing-shape control.Dryden Flight Research Center, Edwards, California

The Motion Base Shuttle MissionSimulator (MBSMS) Strain System is aninnovative engineering tool used tomonitor the stresses applied to theMBSMS motion platform tilt pivotframes during motion simulations inreal time. The Strain System compriseshardware and software produced by sev-eral different companies. The systemutilizes a series of strain gages, ac-celerometers, orientation sensor, rota-tional meter, scanners, computer, andsoftware packages working in unison.By monitoring and recording the in-

puts applied to the simulator, data canbe analyzed if weld cracks or otherproblems are found during routine sim-ulator inspections. This will help engi-neers diagnose problems as well as aidin repair solutions for both current aswell as potential problems.

The system is located with a line-of-sight to the Motion Base for real-timeon-site monitoring. In addition to localmonitoring, several off-site engineeringcomputers are loaded with software al-lowing the user to remotely log onto theStrain System computer, monitor the sys-

tem, and adjust the software settings asrequired. Additional commercial soft-ware products have been programmedto automate the Strain System software.This removes the typical daily manual in-teraction required by the system to boot,record, stop, and save the resultant mo-tion data for future analysis.

This work was done by David C Huber,Karl G. Van Vossen, Glenn W. Kunkel, andLarry W. Wells of United Space Alliance forJohnson Space Center. Further information iscontained in a TSP (see page 1). MSC24386-1

Strain System for the Motion Base Shuttle Mission SimulatorLyndon B. Johnson Space Center, Houston, Texas


An apparatus, denoted the “retractua-tor” (a contraction of “retracting actua-tor”), was designed to help ensure cleanseparation between the cruise stage andthe entry-vehicle subsystem of the MarsScience Laboratory (MSL) mission. Theretractuator or an equivalent mecha-nism is needed because of tubes that (1)transport a heat-transfer fluid betweenthe stages during flight and (2) are cutimmediately prior to separation of thestages retractuator. The role of the re-tractuator is to retract the tubes, after

they are cut and before separation of thesubsystem, so that cut ends of the tubesdo not damage thermal-protection coatson the entry vehicle and do not con-tribute to uncertainty of drag and conse-quent uncertainty in separation velocity.

The retractuator was conceived as aless massive, less bulky, and more power-ful alternative to a traditional spring-ac-tuated retractor. The retractuator is amodified version of a prior pyrotechni-cally actuated cutter. The modificationsinclude alterations of the geometries of

pyrotechnic charges, piston, and cylin-der; replacing the cutter blade with apush rod; and other changes to reduceweight, arrest the piston at the end of itsstroke, and facilitate installation.

This work was done by John C. Gallon,Richard G. Webster, Keith D. Patterson, andMatthew A. Orzewalla of Caltech, Eric T.Roberts of Raytheon Co., and Andrew J.Tuszynski of Columbus Technologies andServices, Inc. for NASA’s Jet Propulsion Lab-oratory. For more information, contact [email protected]. NPO-45680

Pyrotechnic Actuator for Retracting Tubes Between MSL SubsystemsNASA’s Jet Propulsion Laboratory, Pasadena, California

beam curvature equation in closedforms to yield slope and deflectionequations for each domain in recur-sion formats. The final deflection equa-tions in summation forms (called Kodisplacement transfer functions),which contain no structural properties(such as bending stiffness), were thenexpressed in terms of domain length,beam depth factor, and surface bend-ing strains at the domain junctures. Infact, the effect of the structural proper-ties is absorbed by surface strains.

For flying wing structures, the two-linestrain-sensing system is a powerfulmethod for simultaneously monitoringthe bending and cross sectional rota-tions. The two-line strain-sensing systemeliminates the need for installing theshear strain sensors to measure the sur-face distortions through which the wingstructure cross sectional rotations couldbe determined.

The Ko displacement theory com-bined with onboard fiber-optic strain-sensing system forms a powerful tool

for in-flight deformed shape monitor-ing of flexible wings and tails, such asthose often employed on unmannedflight vehicles by the ground-basedpilot for maintaining safe flights. In ad-dition, the real-time wing shape moni-tored could then be input to the air-craft control system for aero-elasticwing-shape control.

This work was done by William L. Ko ofDryden Flight Research Center. Further infor-mation is contained in a TSP (see page 1).DRC-006-024


Physical Sciences

Surface-enhanced x-ray fluorescence(SEn-XRF) spectroscopy is a form of sur-face-enhanced spectroscopy that wasconceived as a means of obtaininggreater sensitivity in x-ray fluorescence(XRF) spectroscopy. As such, SEn-XRFspectroscopy joins the ranks of suchother, longer-wavelength surface-en-hanced spectroscopies as those based onsurface-enhanced Raman scattering(SERS), surface-enhanced resonanceRaman scattering (SERRS), and surface-enhanced infrared Raman absorption(SEIRA), which have been described inprevious NASA Tech Briefs articles.

XRF spectroscopy has been used inanalytical chemistry for determining theelemental compositions of small sam-ples. XRF spectroscopy is rapid andquantitative and has been applied to avariety of metal and mineralogical sam-ples. The main drawback of XRF spec-troscopy as practiced heretofore is thatsensitivity has not been as high as re-quired for some applications.

In SEn-XRF as in the other surface-en-hanced spectroscopies, one exploits sev-eral interacting near-field phenomena,occurring on nanotextured surfaces, thatgive rise to local concentrations of inci-dent far-field illumination. In this case,the far-field illumination comes from anx-ray source. Depending on the chemicalcomposition and the geometry of a givennanotextured surface, these phenomenacould include the lightning-rod effect(concentration of electric fields at thesharpest points on needlelike surfacefeatures), surface plasmon resonances,and grazing incidence geometric effects.In the far field, the observable effect ofthese phenomena is an increase in theintensity of the spectrum of interest — inthis case, the x-ray fluorescence spec-trum of chemical elements of interestthat may be present within a surfacelayer at distances no more than a fewnanometers from the surface.

In experiments, SEn-XRF was demon-strated on aluminum substrates, the sur-faces of some of which had been ran-domly nanotextured (see Figure 1) bybriefly etching them in hydrochloric

Surface-Enhanced X-Ray FluorescenceXRF spectra can be enhanced by large factors.NASA’s Jet Propulsion Laboratory, Pasadena, California

Figure 1. This Atomic-Force-Microscope Image shows a portion of the surface of an aluminum sub-strate that was roughened by a 1-minute etch in concentrated hydrochloric acid.

Figure 2. Silicon X-Ray Fluorescence Peaks at 1.74 keV were measured on aluminum surfaces coatedwith 340-nm-thick layers of silicone oil. The peak from the nanotextured surface was about 2.5 timesas high as that from the nanotextured surface. Taking account of calculations showing that the en-hancement occurs only within the 5-nm-thick sublayer closest to the surface, it was estimated that ifthe surfaces were coated to only 5-nm thickness, the factor of enhancement would be at least 170.

300

200

100

0

0

250

500

750

750500

2500

150

75

0

Note: All dimensionsare in nanometers

1.0 2.01.74

AluminumPeak

SiliconPeak

Spectrum FromNanotextured Surface

Spectrum FromNon-Nanotextured Surface

Photon Energy, keV

Spec

tral

Inte

nsi

ty, U

nsp

ecif

ied

Un

its


acid. Thin layers containing elements ofinterest (Si, I, K, Hg, and Pb) were de-posited on the substrate, variously, fromdilute salt solutions (in the case of K,Hg, and Pb), by vapor sublimation (inthe case of I), or in a thin film of siliconeoil (in the case of Si). XRF spectra of thethus-coated substrate surfaces were ob-tained by use of a commercial XRF mi-croprobe instrument. In some cases, theXRF spectra of elements of interest onthe nanotextured substrates were found

to be enhanced significantly over thespectra from the corresponding sub-strates that had not been nanotextured(for example, see Figure 2).

This work was done by Mark Anderson ofCaltech for NASA’s Jet Propulsion Laboratory.Further information is contained in a TSP(see page 1).


Innovative Technology Assets ManagementJPLMail Stop 202-2334800 Oak Grove DrivePasadena, CA 91109-8099(818) 354-2240E-mail: [email protected] to NPO-44351, volume and number


The idea of designing a microfluidicchannel to slope upward along the di-rection of flow of the liquid in the chan-nel has been conceived to help preventtrapping of gas bubbles in the channel.In the original application that gave riseto this idea, the microfluidic channelsare parts of micro-capillary elec-trophoresis (microCE) devices undergo-ing development for use on Mars in de-tecting compounds indicative of life. It isnecessary to prevent trapping of gasbubbles in these devices because unin-

terrupted liquid pathways are essentialfor sustaining the electrical conductionand flows that are essential for CE. Theidea is also applicable to microfluidic de-vices that may be developed for similarterrestrial microCE biotechnological ap-plications or other terrestrial applica-tions in which trapping of bubbles in mi-crofluidic channels cannot be tolerated.

A typical microCE device in the origi-nal application includes, among otherthings, multiple layers of borosilicatefloat glass wafers. Microfluidic channels

are formed in the wafers, typically by useof wet chemical etching. The figurepresents a simplified cross section ofpart of such a device in which the CEchannel is formed in the lowermostwafer (denoted the channel wafer) and,according to the present innovation,slopes upward into a via hole in anotherwafer (denoted the manifold wafer)lying immediately above the channelwafer. Another feature of the present in-novation is that the via hole in the man-ifold wafer is made to taper to a wider

Slopes To Prevent Trapping of Bubbles in Microfluidic Channels This idea helps to ensure functionality of micro-capillary electrophoresis devices. NASA’s Jet Propulsion Laboratory, Pasadena, California

Since 2006, NASA’s Dryden Flight Re-search Center (DFRC) and Ames Re-search Center have been perfecting anddemonstrating a new capability for ge-olocation of wildfires and the real-timedelivery of data to firefighters. Managedfor the Western States Fire Mission, theAmes-developed Autonomous ModularScanner (AMS), mounted beneath awing of DFRC’s MQ-9 Ikhana remotelypiloted aircraft, contains an infraredsensor capable of discriminating tem-peratures within 0.5 °F (≈ 0.3 °C), up to1,000 °F (≈ 540 °C).

The AMS operates like a digital cam-era with specialized filters to detect lightenergy at visible, infrared, and thermalwavelengths. By placing the AMS aboardunmanned aircraft, one can gather in-

formation and imaging for thousands ofsquare miles, and provide critical infor-mation about the location, size, and ter-rain around fires to commanders in thefield. In the hands of operational agen-cies, the benefits of this NASA researchand development effort can support na-tionwide wildfire fighting efforts. Thesensor also provides data for post-burnand vegetation regrowth analyses.

The MQ-9 Unmanned Aircraft System(UAS), a version of the Predator-B, canoperate over long distances, staying aloftfor over 24 hours, and controlled via asatellite-linked command and controlsystem. This same link is used to deliverthe fire location data directly to fire inci-dent commanders, in less than 10 min-utes from the time of overflight. In the

current method, similarly equippedshort-duration manned aircraft, withlimited endurance and range, mustland, hand-carry, and process data, andthen deliver information to the firefight-ers, sometimes taking several hours inthe process. Meanwhile, many fireswould have moved over great distancesand changed direction. Speed is critical.The fire incident commanders must as-sess a very dynamic situation, and taskresources such as people, ground equip-ment, and retardant-dropping aircraft,often in mountainous terrain obscuredby dense smoke.

This work was done by Mark Pestana ofDryden Flight Research Center. Further infor-mation is contained in a TSP (see page 1).DRC-010-020

Infrared Sensor on Unmanned Aircraft Transmits Time-CriticalWildfire DataThe sensor detects light at visible, infrared, and thermal wavelengths.Dryden Flight Research Center, Edwards, California


opening at the top to further reduce thetendency to trap bubbles.

At the time of reporting the informa-tion for this article, an effort to identifyan optimum technique for forming theslope and the taper was in progress. Ofthe techniques considered thus far, theone considered to be most promising isprecision milling by use of femtosecondlaser pulses. Other similar techniquesthat may work equally well are precisionmilling using a focused ion beam, or asmall diamond-tipped drill bit.

This work was done by Harold F. Greer,Michael C. Lee, J. Anthony Smith, and PeterA. Willis of Caltech for NASA’s Jet PropulsionLaboratory. For more information, [email protected]. NPO-45934

Flow

Channel Wafer

CE Channel

Manifold Wafer

Fluidic Wafer

Tapered Via Hole

The slope of the CE Channel and the taper of the via hole in the manifold wafer help to prevent trap-ping of bubbles.

National Aeronautics andSpace Administration

technology focus electronics/computers - nasa€¦ · tripler at 350-mw input power, and 27-mw...

Documents