Surface Acoustic Wave (SAW) Wireless Passive RF Sensor System

Tutorial

Donald C. Malocha

Department of Electrical Engineering & Computer Science

University of Central Florida

Orlando, Fl. 32816-2450

donald.malocha@ucf.edudonald.malocha@ucf.edu

Piezoelectric Substrate

f1 f4 f6 f0f2 f5 f3

1

Don MalochaUniversity of Central Florida

• Don Malocha, Professor, University of Central Florida• BS, MS and PhD, Univ. of Illinois, UIUC• Texas Instruments, Corporate Research Laboratory, Dallas, MTS• Sawtek, Orlando, Mgr. of Advanced Product Development• Motorola, Visiting/Member of the Technical Staff, Phoenix and Ft.

Lauderdale• Visiting Faculty, ETH, Switzerland, and Univ. of Linz, Austria• Past President, IEEE Ultrasonics, Ferroelectrics and Frequency

Control Society

• WEB site: http://caat.engr.ucf.edu/UCF – nations 2nd largest university

2

3

Acknowledgment•The author wishes to thank continuing support from everyone who has aided us at NASA, and especially Dr. Robert Youngquist, NASA-KSC.

•The foundation of this work was funded through NASA Graduate Student Research Program Fellowships, the University of Central Florida – Florida Space Grant Consortium, and NASA STTR and SBIR contracts.

•Continuing research is funded through NASA STTR/SBIR contracts and industrial collaboration with our industrial partner

Mnemonics Inc. (MNI), Melbourne, Fl.

General Background

4

Multi-Sensor TAG Approaches• Silicon RFID – integrated or external sensors

– Requires battery, energy scavenging, or transmit power– Radiation sensitive– Limited operating temperature & environments

• SAW RFID Tags - integrated or external sensors

– Passive – powered by interrogation signal

– Radiation hard

– Operational temperatures ~ 0 - 500+ K• Resonator – coding in frequency• CDMA- time coding, 40-60 dB loss, wideband• OFC - time & frequency coding, 6-20 dB loss, ultra wide band

5

Why Use SAW Sensors and Tags?• Frequency/time are measured with greatest

accuracy compared to any other physical measurement (10-10 - 10-14).

• External stimuli affects device parameters (frequency, phase, amplitude, delay)

• Operate from cryogenic to >1000oC• Ability to both measure a stimuli and to

wirelessly, passively transmit information• Frequency range (practical) ~100 MHz – 3 GHz• Monolithic structure fabricated with current IC

photolithography techniques, small, rugged

6

7

What is a typical SAW Device?• A solid state device

– Converts electrical energy into a mechanical wave on a single crystal substrate

– Provides very complex signal processing in a very small volume

• Approximately 4-5 billion SAW devices are produced each year

Applications:Cellular phones and TV (largest market)

Military (Radar, filters, advanced systems

Currently emerging – sensors, RFID

SAW Principle - Piezoelectricity

Squeezing a piezo-crystal creates a voltage.A voltage can compress or dilate a piezo-crystal.

8

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5-80

-70

-60

-50

-40

-30

-20

-10

Time (s)

dB

(s

21)

Direct SAW response

Reflector response

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5-80

-70

-60

-50

-40

-30

-20

-10

Time (s)

dB

(s

21)

Direct SAW response

Reflector response

SAW BasicsTransduction & Reflection fro SAW Sensors

20λ0 50λ0 50λ020λ0 20λ0 50λ0 50λ020λ0

SAW - mechanical wave trapped to the surface

Transduction via piezoelectric effect

Velocity ~ 3000 - 4000 m/sec

Wavelength @ 1 GHz ~ 3 um

Line resolution at 1 GHz ~ .75 um

Reflection via Bragg reflector structure

Bragg reflector

DC Effect

RF to SAW

9

10

SAW Materials to Meet Sensor Needs

Material Crystal cutCoupling

coefficientTemperature coefficient

SAW Velocity

Max Temp

LiNbO3 Y,Z 4.6% 94 ppm/ºC 3488 m/s ~500 ºC

128ºY,X 5.6% 72 ppm/ºC 3992 m/s ~500 ºC

LiTaO3 Y,Z 0.74% 35 ppm/ºC 3230 m/s ~500 ºC

Quartz ST 0.16% 0 ppm/ºC 3157 m/s 550 ºC

Langasite Y,X 0.37% 38 ppm/ºC 2330 m/s >1000 ºC

138ºY,26ºX 0.34% ~0 ppm/ºC 2743 m/s >1000 ºC

SNGS Y,X 0.63% 99 ppm/ºC 2836 m/s >1000 ºC

SAW travels ~ 105 slower than EM waveSAW wavelength @ 1 GHz ~ 3 um

SAW/IC Fabrication Techniques

The dark line in each micrograph is a 23 um gold wire

SAW reflector gratings

SAW Transducer

Lines are ~ .8 um

SAW reflector gratings

• SAW devices @ 1 GHz require submicron lithography.• Standard IC thin films, photolithography and processing are used. 11

Basic Passive Wireless SAW System

Sensor 3

Sensor 1

Sensor 2

Clock

Interrogator

Post Processor

12

Goals:•Interrogation distance: 1 – 50 meters •# of devices: 10’s – 100’s - coded and distinguishable at TxRx•Aerospace applications – rad hard, wide temp., solid state, etc.•Single platform and TxRx for differing sensor combinations

Sensor #1 Gas

Sensor #3 Temperature

Sensor #2 Pressure

RFID and SAW Introduction

13

RFID Sensor

• RFID Acquisition– Priority for system– Coding approach– Demodulation

approach– System Parameters

• Measurand Extraction– RFID is acquired– S/N ratio– Accuracy– Acquisition rate

Two primary system functions: RFID and extraction of the measurand. The RFID must first be acquired and then the measurand extracted. The presentation will address these issues for a temperature sensor system.

14

Diversity for Identification

• Frequency Spectrum Diversity per Device– Coding– Divide into frequency bands

• Time Delay per Device– Different offset delays per device– Pulse position modulation– Time allocations minimize code collisions

• Spatial Diversity – device placement

• Sensor & Tx-Rx Antenna Polarization

• Use combinations of all to optimize system15

16

• One port devices return the altered interrogation signal

• Range depends on embodiment • Range increased using coherent

integration of multiple responses• Interrogator used to excite devices• Several embodiments are shown next

Brief Introduction to Wireless SAW Sensors

17

Reflective Delay Line Sensor

• First two reflectors define operating temperature range of the sensor

• Time difference between first and last echoes used to increase resolution of sensor

• No coding as shown

“Wireless Interrogator System for SAW-Identification-Marks and SAW-Sensor Components”,

F. Schmidt, et al, 1996 IEEE International Frequency Control Symposium

18

SAW Chirp Sensor

• Increased sensitivity when compared with simple reflective delay line sensor

• Multi-sensor operation not possible due to lack of coding

“Spread Spectrum Techniques for Wirelessly Interrogable Passive SAW Sensors”,

A. Pohl, et al, 1996 IEEE Symposium on Spread Spectrum Techniques and Applications

19

Impedance SAW Sensors

• External classical sensor or switch connected to second IDT which operates as variable reflector

• Load impedance causes SAW reflection variations in magnitude and phase

• No discrimination between multiple sensors as shown

“State of the Art in Wireless Sensing with Surface Acoustic Waves”,

W. Bulst, et al, IEEE UFFC Transactions, April 2001

SAW RFID Practical Approaches

• Resonator– Fabry-Perot Cavity– Frequency selective, SAW device Q~10,000

• Code Division Multiple Access (CDMA)– Delay line – single frequency Bragg reflectors– Pulse position encoding

• Orthogonal Frequency Coding (OFC)– Delay line, multi-frequency Bragg reflectors– Pulse position encoding– Frequency coupled with time diversity

20

SAW Resonator

D D

Grating GratingIDT

354.6 354.8 355 355.2 355.4 355.6 355.8 356 356.2 356.4-14

-12

-10

-8

-6

-4

-2

Frequency, MHz

S11

mag

nitu

de (

dB)

experimentalpredicted

“Remote Sensor System Using Passive SAW Sensors”,

W. Buff, et al, 1994 IEEE International Ultrasonics Symposium

Q~10,000

• Resonant cavity• Frequency with maximum returned

power yields sensor temperature• High Q, long time response• Coding via frequency domain by

separating into bands

21

SAW CDMA Delay Line

CDMA Tag Concept

•Single frequency Bragg reflectors

•Coding via pulse position modulation

•Large number of possible codes

•Short chips, low reflectivity - (typically 40-60 dB IL)

•Early development by Univ. of Vienna, Siemens, and others

22

CDMA Tag

SAW OFC Delay Line

OFC Tag

•Multi-frequency (7 chip example)

•Long chips, high reflectivity

•Orthogonal frequency reflectors –low loss (6-10 dB)

•Example time response (non-uniformity due to transducer)

OFC Tag

DUT - RF probe connected to transducer

Bragg reflector gratings at differing frequencies

Micrograph of device under test (DUT)

23

DiscussionResonator, CDMA, and OFC embodiments have all been successfully demonstrated and applied to various applications. Devices and systems have been built in the 400 MHz, 900 MHz and 2.4 GHz bands by differing groups.

Resonator•Minimal delay•Narrowband PG~1•Fading•Frequency domain coding•High Q – long impulse response•Low loss sensor

CDMA•Delay as reqd. ~ 1usec•Spread Spectrum

Fading immunityWideband PG >1

•Time domain coding•Large number of codes using PPM

OFC•Delay as reqd. ~ 1usec•Spread Spectrum

Fading immunityUltra Wide Band PG >>1

•Time & frequency domain coding•Large number of codes using PPM and diverse chip frequencies

24

OFC Sensor Embodiment

25

SAW OFC Sensor Introduction Conventional wisdom at the time:• “ Orthogonality in frequency is not feasible with coded reflective

passive SAW sensors.”, “Spread Spectrum Techniques for Wireless Interrogable Passive SAW Sensors”, A. Pohl, et. al., IEEE 4th International Symposium on Spread Spectrum Techniques and Applications, 1996, pp. 730.

• “D. Malocha and coworkers recently developed Orthogonal Frequency Coding for SAW tags [25]. …….This approach can be applied to sensors and for identification of a limited number of sensors, but it can hardly be used for ID tags with large numbers of codes.” Review on SAW RFID Tags, V. P. Plessky, and L.M. Reindl, IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, vol. 57, no. 3, March 2010, pp.654

• First OFC publication by UCF group in 2004 and working system in 2009. The use of spread spectrum frequency and time coding had been overlooked as either not possible or too complicated. For RFID sensors, the approach is both feasible, advantageous, and demonstrated. 26

OFC Historical Development• Chose 1st devices at 250 MHz for feasibility• Several different OFC sensors demonstrated• Demonstrated harmonic operated devices at

456, 915 MHz and 1.6 GHz• Fundamental device operation at 915 MHz• Devices in the +1 GHz range in 2010• First OFC system at 250 MHz• Current OFC system at 915 MHz• First 4 device wireless operation in 2009• Mnemonics demonstrates first chirp OFC

correlator receiver in 201027

Why OFC SAW Sensors?• A game-changing

approach• All advatageous of

SAW technology • Wireless, passive and

multi-coded sensors• Frequency & time offer

greatest coding diversity

• Single communication platform for diverse sensor embodiments

• Radiation hard• Wide operational

temperature range

28

Piezoelectric Substrate

f1 f4 f6 f0f2 f5 f3

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.80

0.2

0.4

0.6

0.8

Normalized Frequency

Mag

nitu

de (

Lin

ear)

Schematic of OFC SAW ID Tag

0 1 2 3 4 5 6 71

0.5

0

0.5

1

Normalized Time (Chip Lengths)

Time domain chips realized in Bragg reflectors having differing carrier frequencies and frequencies are non-sequential which provides coding

Sensor bandwidth is dependent on number of chips and sum of chip bandwidths. Frequency domain of Bragg reflectors: contiguous in frequency but shuffled in time

29

Example 915 MHz SAW OFC Sensor

FFT

US QuarterSAW Sensor

SAW OFC Reflector Chip Code

f4 f3 f1 f5 f2

30

Piezoelectric Substrate

f1 f4 f2 f6 f5f0 f3

100 150 200 250 300 350 400-0.7

-0.6

-0.5

-0.4

-0.3

-0.2

-0.1

0

Frequency (MHz)

S1

1 (

dB

)

OFC Sensor Response

100 150 200 250 300 350 400-0.7

-0.6

-0.5

-0.4

-0.3

-0.2

-0.1

0

Frequency (MHz)

S1

1 (

dB

)

OFC Sensor Response

100 150 200 250 300 350 400-0.7

-0.6

-0.5

-0.4

-0.3

-0.2

-0.1

0

Frequency (MHz)

S1

1 (

dB

)

OFC Sensor Response

100 150 200 250 300 350 400-0.7

-0.6

-0.5

-0.4

-0.3

-0.2

-0.1

0

Frequency (MHz)

S1

1 (

dB

)

OFC Sensor Response

100 150 200 250 300 350 400-0.7

-0.6

-0.5

-0.4

-0.3

-0.2

-0.1

0

Frequency (MHz)

S1

1 (

dB

)

OFC Sensor Response

100 150 200 250 300 350 400-0.7

-0.6

-0.5

-0.4

-0.3

-0.2

-0.1

0

Frequency (MHz)

S1

1 (

dB

)

OFC Sensor Response

100 150 200 250 300 350 400-0.5

-0.45

-0.4

-0.35

-0.3

-0.25

-0.2

-0.15

-0.1

-0.05

0

Frequency (MHz)

S1

1 (

dB

)

OFC Sensor Response

100 150 200 250 300 350 400-0.5

-0.45

-0.4

-0.35

-0.3

-0.25

-0.2

-0.15

-0.1

-0.05

0

Frequency (MHz)

S1

1 (

dB

)

OFC Sensor Response

SAW OFC RFID signal – Target reflection as seen by antenna

S11 w/ absorber and w/o reflectors

31

SAW

absorber

Coded SAW chips are bound in frequency and received sequentially in time

S11 w/o absorber and w/ reflectors

OFC vs CDMA Number of possible codes versus number of chips

for same chip configuration

CDMA: # codes=2N

OFC: # codes=N!*2N where N= #chips 32

Effect of Code Collisions from Multiple SAW RFID Tags -Simulation

0 1 2 3 4 5 6 7 810

0

10

Optimal Correlation OutputActual Recevied Correlation Output

3rd Bit

Time Normalized to a Chip Length

Nor

mal

ized

Am

plitu

de

Due to asynchronous nature of passive tags, the random summation of multiple correlated tags can produce false correlation peaks and

erroneous information

33

34

OFC Coding• Time division diversity (TDD): Extend the possible

number of chips and allow delay and phase modulation– # of codes increases dramatically, M>N chips, >2M*N!– Reduced code collisions in multi-device environment

Sensor #1

0 5 102

1

0

1

2

Time Response

Time Normalized to Chip Length

Norm

ali

zed

Am

pli

tud

e

456 MHZ SAW OFC TDD Coding

35

A 456 MHz, dual sided, 5 chip, tag COM-predicted and measured time responses illustrating OFC-PN-TDD coding. Chip amplitude variations are primarily due to polarity weighted transducer effect and fabrication variation.

1.5 2 2.5 3 3.5-105

-100

-95

-90

-85

-80

-75

-70

-65

-60

-55

Time (s)

s 11 (

dB

)

Simulation

Experiment

OFC FDM Coding• Frequency division multiplexing: System uses N-frequencies

but any device uses M < N frequencies– System bandwidth is N*Bwchip

– OFC Device is M*BWchip• Narrower fractional bandwidth• Lower transducer loss• Smaller antenna bandwidth

36

Sensor #1

Sensor #2

36

Antenna and SAW Sensor Design Considerations

37

The plots show that there is a minimum size at a given frequency to attain a desired fractional bandwidth.As the frequency increases, a larger fractional bandwidth is achievable for a smaller antenna size.As the effective size of the antenna increases, the gain and bandwidth both increase.

SAW Electrically Small Antenna Gain and Bandwidth

38

SAW TARGET – SAW + ANTENNA

UCF Initial Design

250 MHz Disk Monopole Antennas

Large dinner plate design met fractional

bandwidth, but hardly miniature

compared to SAW sensor size

39

Target Gain vs. FrequencyAnalysis points to ~1 GHz

SAW, antenna and net gain in dB, and fractional bandwidth, versus frequency for a 3cm radius ESA. Assumes a SAW propagation length of 5 usec.

where f is in GHz

Good fo region

%BW

40

• Designed on 32mil FR4 (εr=4.7 and tan(δ)=0.015)

• Entire structure optimized in IE3D between 800MHz and 1GHz

Wideband Open-Sleeve Dipole Antenna

4141

SAWtenna @ 915 MHz

Fully integrated on-wafer SAW OFC sensor and antenna

Wireless OFC SAWtenna time domain response

Test wafer-level SAW & antenna integration

42

Miniature 915MHz Integrated OFC SAW-Patch Antenna

43

Synchronous Correlator Transceiver

44

Synchronous Transceiver - Software Radio

SAWsensor

RF Oscillator

Digital control and analysis circuitry

SAW up-chirp filter

SAW down-chirp filter

IF Oscillator

A / D

IF Filter

• Pulse Interrogation: Chirp or RF burst• Correlator Receiver Synchronous• Software Radio Based

915 MHz Pulsed RF Transceiver Block Diagram

45

Temperature ExtractionUsing Adaptive Correlator

Comparison of ideal and measured matched filter of two different SAW sensors : 5-chip frequency(below)

-0.2 -0.15 -0.1 -0.05 0 0.05 0.1 0.15 0.2-30

-25

-20

-15

-10

-5

0

Time (s)

Am

plitu

de (

Nor

mal

ized

)

Experimental

Ideal

-0.2 -0.15 -0.1 -0.05 0 0.05 0.1 0.15 0.2-30

-25

-20

-15

-10

-5

0

Time (s)

Am

plitu

de (

Nor

mal

ized

)

Experimental

IdealNS403

NS401

Normalized amplitude (dB) versus time

Stationary plots represent idealized received SAW sensor RFID signal at ADC. Adaptive filter matches sensor RFID temperature at the point when maximum correlation occurs.

46

Synchronous Correlator Receiver

Block diagram of a correlator receiver using ADC

-0.2 -0.15 -0.1 -0.05 0 0.05 0.1 0.15 0.2-30

-25

-20

-15

-10

-5

0

Time (s)

Am

plitu

de (

Nor

mal

ized

)

Experimental

Ideal

OFC Single Sensor Signal

Correlation Output

Temperature Extraction

47

250 MHz Wireless Pulsed RF OFC SAW System - 2nd Pass

An OFC SAW temperature sensor data run on a free running hotplate from an improved 250 MHz transceiver system. The system used 5 chips and a fractional bandwidth of approximately 19%. The dashed curve is a thermocouple reading and the solid curve is the SAW temperature extracted data. The SAW sensor is tracking the thermocouple very well; the slight offset is probably due to the position and conductivity of the thermocouple.

50 cm 50 cm

30 cm 30 cm

SAW Sensor/Tag

Interrogator(Transmitter)

Receiver

Hot Plate

78°CThermal

Controller

Thermal Couple

48

SAW 915 MHz Correlator Transceiver

49

MNI Transceiver Design

• Pulsed RF Chirp

• Correlator Receiver– Synchronous operation– Integration of multiple “pings”– OFC processing gain

• Adaptive filter temperature extraction

• Software radio based approach for versatility

50

Current Sensor System Results• 915 MHz transceiver developed by Mnemonics,

Inc. (MNI), Melbourne, Fl– RF Chirp 700nsec, 28dBm peak power– Synchronous receiver

• OFC SAW temperature sensors developed by UCF– YZ LiNbO3, 5 chip OFC delay line sensor– 915 MHz fundamental, 0.8 um electrodes

• Correlator software developed at UCF

51

Critical Transceiver Operational Parameters

• EM Path Loss Considerations

• Electrically Small Antennas (ESA)

• SAW Device Propagation Loss

• Target Gain versus Center Frequency

• Integrated SAW and Antenna

52

EM Path Loss versus Range

• EM isotropic two-way path loss for 3 differing operational frequencies: 0.25, 0.5 and 1 GHz - solid lines.

• The dotted traces are the thermal noise levels at 3 differing bandwidths, 25, 72, and 200 MHz.

• Path loss increases @ 40dB/decade w/ increasing range or frequency

53

RF Transceiver:Sensor Overview

• OFC with single wideband transducer

• Center Frequency: 915 MHz

• Bandwidth: Chirp - ~78 MHz

• Number of Chips: 5

• Chip length 54ns/each, total reflector length 270ns

• Substrate: YZ LiNbO3

54

SAW 915 MHz OFC Sensor

• SAW sensor acts as RFID and sensor

• All antenna & transducer effects are doubled

• Antenna gain and bandwidth are dependent on size scaled to frequency

• SAW propagation loss is frequency dependent

55

Parameter Definitions(extensive list of variables)

• PG= signal processing gain of the system (= τ·B)

• PL= path loss• NF= receiver noise figure• Next= external noise source

referenced to antenna output

• NADC= ADC equivalent noise• Nsum= number of

synchronous integrations in ADC

• PGC = pulse compression gain from chirp interrgogation

56

• ADC= ideal analog-to-digital converter

• MDS= minimum detectable signal at ADC

• S= signal power measured at ADC

• N= noise power measured at ADC

• kT= thermal noise energy• EIRP= equivalent radiated

power• GRFIDS= RFIDS gain (less than

unity for passive device)• GRx-ant= gain of the receiver

antenna• GRx= receiver gain from

antenna output to ADC

Range Prediction• For passive RFIDS, the range is given from Friis

equation as Range =r = PL.25·[vEM/(4·π·fo)] ;

where vEM=free space velocity

• A minimum S/N is determined for detection, and the maximum range, in meters, achievable, given in dB, is obtained as

rmax-dB=.25·{GPDL+Gsys+Nsum-[S/Nmin]} -10·log[(4·π·f)/vEM],

where

GPDL=[EIRP/(NF*·kT/τau)] = power-detection level gain and Gsys = [(GRFIDS·GRx-ant)]

57

RF Chirp Transceiver Parameters

• Power to antenna = 30dBm• Pulse-length = 700ns, 20Vpp

• Antenna Gain = 9dB

• Bandwidth = 74MHz

• Receiver Gain = 45dB

• NF = 15dB

• PGC= 49 = 17 dB

58

Range Prediction for MNI Receiver for RFID Detection (not sensor)

• Range is a function of the complete system loop gain, shown in solid line (red). Loop gain is dependent on the transmit power, noise and gain in the system. Typical loop gains are realistically achievable between 100 to 180 dB. The box shows the predicted loop gain for the MNI/UCF system, which is very close to measurements obtained. 59

Chirp Transceiver: SAW OFC Sensor Range Experiment

• Single sensor only; no signal integration

• Multiple distances from 1.2m to 20m

• 0 to 20dB additional attenuation at each step

• 128 readings taken per distance per attenuation

• Longest distance of successful interrogation 7m

• Reading error .07 corresponds to 60% of all data points within 5°C (3.5%)

60

Practical Extension

• NF = 18dB → 8dB (∆G = 10dB)

• GSAW = -23dB → -10dB (∆G = 13dB)

• GPSI = 12dB → 22dB (∆G = 10dB)

• Total improvement: 33dB

• Approximately extended range: 80m61

915 MHz OFC Temperature Sensor SystemMeasured Device Data in a Hallway

Data is measured in a hallway approximately 2.1 meters wide. Antennas: transmit is a wideband 1 dB dipole; receive is a 9 dB Yagi. The system loop gain is calculated at ~40 dB (+/-3 dB). Transmit signal is a single, 700 nsec, 915 MHz chirp pulse. The OFC SAW device uses 5 chips, each with an approximate 15 MHz bandwidth. SAW device processing gain is 25. Slope of the fit measured data is -38.7 dB/decade; close to the 40 dB/decade expected for isotropic radiation path loss. The hallway is probably producing a waveguiding effect and external noise was low during testing. Test shows that some environments can produce long ranges.

62

OFC SAW Correlator Receiver Tag Ranging

• Distance from interrogator to the sensor can be extracted based on EM delay (8m per chip length – 54ns)

• X-axis indicates various distances at which sensor was placed away from interrogator

• Cross-marks indicate distance from interrogator on y-axis

• 128 Measurements were made for each step

• Blue box indicates spread of a half of all data

• Black boundaries indicate spread of 99.3% of all data

• Red pluses indicate outliers

63

UCF Sensor Development• The following are a few of

the successful UCF sensor projects

• The aim is to enable wireless acquisition of the sensors data

• The further goal is to develop a multi-sensor system for aerospace applications

• Successful wireless sensing has been demonstrated for temperature, liquid, closure, and range

• There is an extensive body of knowledge on sensing

• Wired SAW sensing has quite an extensive body of knowledge and continues

• Wireless SAW sensing has been most successfully demonstrated for single, or very few devices and in limited environments

64

UCF OFC Sensor Successful Demonstrations

• Temperature sensing– Cryogenic: liquid nitrogen – Room temperature to 250oC– Currently working on sensor for operation to

750oC

• Cryogenic liquid level sensor: liquid nitrogen

• Pressure/Strain sensor• Hydrogen gas sensor• Closure sensor with temperature 65

Four-sensor operation

• Four OFC SAW sensors are co-located in close range to each other at a distance of 0.8m to 1.2m

• Sensors NS402 and NS404 remained at room temperature

• Sensor NS401 heated to 140°C

• Sensor NS403 cooled to -130°C

• Data was taken simultaneously from all four sensors and then temperature extracted in the correlator receiver software

• Error is within ±5°C (±3.5% for given dynamic range) 66

Differential SAW OFC Thin Film Gas Sensor Embodiment

2.00 mm

1.25 mm 1.38 mm 1.19 mm2.94 mm

6.75 mm

f3 f5 f0 f6 f2 f4 f1

Piezoelectric Substrate

f3 f5 f0 f6f2 f4 f1

f1 f4 f2 f6f0 f5 f3

67

68

Temperature Sensor using Differential Delay Correlator Embodiment

Piezoelectric Substrate

f1 f4 f6 f0f2 f5 f3f1f4f6f0 f2f5f3

Temperature Sensor Example

250 MHz LiNbO3, 7 chip reflector, OFC SAW sensor tested using temperature controlled RF probe station

69

Temperature Sensor Results

• 250 MHz LiNbO3, 7 chip reflector, OFC SAW sensor tested using temperature controlled RF probe station

• Temp range: 25-200oC• Results applied to simulated

transceiver and compared with thermocouple measurements

0 20 40 60 80 100 120 140 160 180 2000

20

40

60

80

100

120

140

160

180

200Temperature Sensor Results

Time (min)

Te

mp

era

ture

( C)

LiNbO3 SAW Sensor

Thermocouple

70

OFC Cryogenic Sensor Results

0 5 10 15 20 25-200

-150

-100

-50

0

50

Time (min)

Tem

pera

ture

( C

)

ThermocoupleLiNbO

3 SAW Sensor

Scale

Vertical: +50 to -200 oC

Horizontal: Relative time (min)

Measurement system with liquid nitrogen Dewar and vacuum chamber for DUT

OFC SAW temperature sensor results and comparison with thermocouple measurements at cryogenic temperatures. Temperature scale is between +50 to -200 oC and horizontal scale is relative time in minutes.

71

Schematic and Actual OFC Gas Sensor

Piezoelectric Substrate

f1 f0f2 f3f1f0 f2f3

•For palladium hydrogen gas sensor, Pd film is in only in one delay path, a change in differential delay senses the gas (τ1 = τ2) (in progress)

Differential mode OFC Sensor Schematic

Actual device with RF probe

Hydrogen Gas sensor Palladium Background Information

• The bulk of PD research has been performed for Pd in the 100-10000 Angstrom thickness

• Morphology of ultra-thin films of Pd are dependent on substrate conditions, deposition and many other parameters

• Pd absorbs H2 gas which causes lattice expansion of the Pd film – called Hydrogen Induced Lattice Expansion (HILE) – Resistivity reduces

• Pd absorbs H2 gas which causes palladium hydride formation – Resistivity increases

• Examine these effects for ultra-thin films (<5nm) on SAW devices

HILE - Each small circle represents a nano-sized

cluster of Pd atoms

CO

NTA

CT

CO

NTA

CT

W ithout H2

CO

NTA

CT

CO

NTA

CT

With H2

72

Pd Films on SAW DevicesSchematic of Test Conditions

• Control: SAW delay line on YZ LiNbO3 wafers w/ 2 transducers and reflector w/o Pd film

• Center frequency 123 MHz

• (A) SAW delay line w/ Pd in propagation path between transducer and reflector

• (B) SAW delay line w/ Pd on reflector only

Pd Film

(A )

(B )

Pd

Film

1.27 mm

73

Hydrogen Gas Sensor Results:

2% H2 gas

74

1.7 1.8 1.9 2 2.1 2.280

76

72

68

64

60

56

52

48

44

40

36

32

28

24

20

Delay Line w/o PdAfter Pd FilmDuring 1st H2 ExposureAfter 1st H2 ExposureDuring 2nd H2 ExposureAfter 2nd H2 ExposureDuring 3rd H2 ExposureAfter 3rd H2 ExposureDuring 4th H2 ExposureAfter 4th H2 Exposure

Time (micro-seconds)

Nor

mal

ized

Mag

nitu

de (

dB)

Pd

Film

100 1 103

1 104

1 105

0

40

80

120

160

200

240

3410

3425

3440

3455

3470

3485

3500

Loss/cm @ 123 MHzLoss/cm due to Pd FilmLoss/cm due to Pd Film After Final H2 Gas ExposureLoss/cm due to successive H2 exposureSAW VelocitySAW Velocity due to Pd FilmSAW Velocity due to Pd Film After Final H2 Gas ExposureSAW Velocity due to successive H2 exposure

Propagation Loss (dB/cm) and Velocity(m/s) vs. Film Resistivity

Resistivity (ohm-cm)

Los

s (d

B/c

m)

SA

W V

eloc

ity (

m/s

)

Pd

Film

Nano-Pd Film – 25 Ang.

•The change in IL indicates a <20 dB sensitivity range and further tests were < 50 dB!

•Sensitive hydrogen sensor is possible.

Theory (lines) versus measurement data

Cantilever Sensor Results

• Initial cantilever sensor results

• Apply results to strain and pressure sensors

75

OFC Cantilever Strain Sensor

• Measure Delay versus Strain

76

Plot generated by ANSYS demonstrating the strain distribution along the z-axis of the crystal.

Test fixture, this shows the surface mount package, which contains the cantilever device, securely clamped down onto a PC board which is connected to a Network Analyzer.

OFC Cantilever Strain Sensor

77

78

Applications

• Current efforts include OFC SAW liquid level, hydrogen gas, pressure and temperature sensors

• Multi-sensor spread spectrum systems• Cryogenic sensing• High temperature sensing• Space applications• Turbine generators• Harsh environments• Ultra Wide band (UWB) Communication

– UWB OFC transducers• Potentially many others

Current to Future

79

Vision for Future• Multiple access, SAW RFID sensors• SAW RFID sensor loss approaching 6 dB

– Unidirectional transducers– Low loss reflectors

• New and novel coding• New and novel sensors• New materials for high temperature

(1000oC) and harsh environments• SAW sensors in test space flight and

support operations in 1 to 5 years80

Ultra Wide Band

• Code coliision reduction

• Multi-bands for multi-sensors

• Subsets of sensors activated at any given time

• Narrower band antennas, lower loss devices

81

• BW defined by chirp, not by individual sensors

• Could use a frequency hopped chirp system

• Frequency diversity is increased

SAW Research at UCF• UCF Center for Acoustoelectronic

Technology (CAAT) has been actively doing SAW and BAW research for over 25 years

• Research includes communication devices and systems, new piezoelectric materials, & sensors

• Capabilities include SAW/BAW analysis, design, mask generation, device fabrication, RF testing, and RF system development

• Current group has 6 PhDs & 1 Post-doc82

Capabilities

• Proprietary software: COM analysis & design, parameter extraction, data acquisition and test

• UCF device fabrication to < .8um resolution• In-house mask fab & thin film capabilities• Complete RF SAW characterization facility• Extensive RF laboratory for system development 83

Research Areas

Thin Films

Processing

Material Charaterization

Measurement

SensorsDesign & Analysis

Center forApplied

AcoustoelectronicsTechnology

Device/SystemFabrication

Synthesis

Modeling

UCF SAW OFC Contracts & Intellectual PropertyA. 6 – Phase I and 4 –Phase II STTR/SBIRs on SAW OFC Sensors

B. NASA KSC, Langley, and JSC contracts

C. Fellowships from NASA, NSF, Motorola, NSDEG, UCF, McKnight, and Florida Space Grant

D. Patents on SAW OFC:

#7,642,898 D.C. Malocha and Puccio, Orthogonal Frequency Coding for Surface Acoustic Wave Communications, Tag, and Sensors, Jan. 5, 2010.

#7,623,037 D.C. Malocha, Multi-transducer/antenna surface acoustic wave device sensor and tag, November 24, 2009.

#7,825,805, D.C. Malocha and D. Puccio, Delayed Offset Multi-Track OFC Sensors and Tags, Nov. 2, 2010.

#7,777,625, D.C. Malocha and D. Puccio, Weighted Reflectors for OFC Coding, Aug. 17, 2010.

#7,791,249, D.C. Malocha and N.Y. Kozlovski, SAW Coding for OFC Devices,

Appl # 12,618,034, D.C. Malocha and N. Kozlovski, Coding for Surface Acoustic Wave Devices, Filed Nov. 13, 2009.

Several in process84

Conclusion

• 915 MHz OFC SAW temperature sensor system has been demonstrated

• Current tests show 10 meter open range• 4 sensors have been simultaneously

interrogated and measured• Range predictions and measured data have

been shown• Wireless passive SAW sensors are a

“game-changing” technology

85