High Impedance Amplifiers for Non-Contact Bio-Potential Sensing

161 Pages·2013·17.17 MB·English

High Impedance Amplifiers for
Non-Contact Bio-Potential
Sensing
by
Brett Ryan
Athesis
submittedtotheVictoriaUniversityofWellington
infulfilmentofthe
requirementsforthedegreeof
MasterofEngineering
inElectronicandComputerSystems.
VictoriaUniversityofWellington
2013

Abstract
This research develops a non-contact bio-potential sensor which can quickly re-
spond to input transient events, is insensitive to mechanical disturbances, and op-
erates with a bandwidth from 0.04Hz – 20kHz, with input voltage noise spectral
√
densityof200nV/ Hzat1kHz.
Initial investigations focused on the development of an active biasing scheme
to control the sensors input impedance in response to input transient events. This
scheme was found to significantly reduce the settling time of the sensor; however
the input impedance was degraded, and the device was sensitive to distance fluc-
tuations. Furtherresearchwasundertaken,andacircuitdevelopedtopreservefast
settlingtimes,whilstdecreasingthesensitivitytodistancefluctuations.
Anovelamplifierbiasingnetworkwasdevelopedusingapairofjunctionfield
effect transistors (JFETs), which actively compensates for DC and low frequency
interference, whilst maintaining high impedance at signal frequencies. This bi-
asing network significantly reduces the settling time, allowing bio-potentials to
be measured quickly after sensor application, and speeding up recovery when the
sensorisinsaturation.
Further work focused on reducing the sensitivity to mechanical disturbances
even further. A positive feedback path with low phase error was introduced to re-
ducetheeffectiveinputcapacitanceofthesensor. Tuningofthepositivefeedback
loopgainwasachievedwithcoarseandfinecontrolpotentiometers,allowingvery
precise gains to be achieved. The sensor was found to be insensitive to distance
fluctuationsofupto0.5mmat1Hz,andupto2mmat5kHz.
As a complement to the non-contact sensor, an amplifier to measure differen-
tialbio-potentialswasdeveloped. ThisdifferentialamplifierachievedaCMRRof
greaterthan100dBupto10kHz. Precisefixedgainsof20±0.02dB,40±0.01dB,
60±0.03dB, and 80±0.3dB were achieved, with input voltage noise density of
√
15nV/ Hzat1kHz.
Acknowledgments
Without the assistance and support of others this research would not have been
possible. I would like to thank my supervisor, Dr. Paul Teal for giving me the
opportunitytopursuethisfascinatingsubject,andfortheendlesssupportanden-
couragement he provided. Eberhard Deuss, for his initial guidance on designing
high impedance amplifiers. Sam Turner for introducing me to the project, and
transferring his enthusiasm for the project to me. Tim Exley for his criticism,
healthyskepticism,andfortakingthetimetodiscusstechnicalproblemswithme.
JasonEdwardsandSeanAndersonforsupplyingmewiththetoolsIneeded,when
Ineededthem. ManuPouajen-BlakistonandNickGrinterfortheiramazingcraft-
manshipontheshieldedtestenclosure,andsensorenclosure. Myofficematesfor
tolerating my often sprawling collection of test rigs, papers, tools, and texts, and
for shutting the door quietly when I was performing sensitive measurements. To
my technically minded friends, in particular Ihab Sinno, Jim Murphy and Patrick
Herd,thankyouforindulginginmytechnicalproblems,talkingtoyouhelpedme
work through a lot of technical issues. To my friends, particularly Mohammad
Zareei, thank you for providing distractions from my work, and encouraging me
whenIwasfeelingdespondent. Tomyflatmates,thanksfornotburningthehouse
down, also Sam Geers, your friendship in the final weeks was outstanding, your
humour, creativity and nutritional support dragged me across the finish line. To
John and the team at milk and honey cafe, thanks for the delicious strong coffee,
humour, and scones. To my parents, thank you for everything! I love you guys
so much and your encouragement has been invaluable. Lastly I’d like to thank
my best friend Kimberley Berends, your encouragement, criticism, financial and
iii
iv
emotional support have not only been essential to the completion of this thesis,
buttotheentiretyofmyuniversitycareer.
Contents
1 Introduction 1
1.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 ProblemDefinition . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 ResearchGoals . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2 Background 5
2.1 Bioelectricity . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.2 HistoryofElectrophysiology . . . . . . . . . . . . . . . . . . . . 5
2.3 ContactSensors . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.4 Non-ContactSensors . . . . . . . . . . . . . . . . . . . . . . . . 9
2.5 OperationalAmplifierImperfections . . . . . . . . . . . . . . . . 13
3 LaboratoryEnvironment 17
3.1 Shielding,GuardingandGrounding . . . . . . . . . . . . . . . . 17
3.2 ShieldedTestenvironment . . . . . . . . . . . . . . . . . . . . . 33
4 Methodology 35
4.1 InputCapacitanceMeasurement . . . . . . . . . . . . . . . . . . 35
4.2 InputBiasCurrentMeasurement . . . . . . . . . . . . . . . . . . 37
4.3 TransferFunctionMeasurements . . . . . . . . . . . . . . . . . . 41
4.4 NoiseMeasurements . . . . . . . . . . . . . . . . . . . . . . . . 45
4.5 GainvsDistanceCharacterisation . . . . . . . . . . . . . . . . . 47
v
vi CONTENTS
5 DifferentialAmplificationSystem 51
5.1 Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
5.2 DesignandEvaluation . . . . . . . . . . . . . . . . . . . . . . . 54
5.2.1 SystemOverview . . . . . . . . . . . . . . . . . . . . . . 54
5.2.2 InstrumentationAmplifier . . . . . . . . . . . . . . . . . 55
5.2.3 Anti-AliasingFilters . . . . . . . . . . . . . . . . . . . . 61
5.2.4 HighGainAmplifier . . . . . . . . . . . . . . . . . . . . 64
5.2.5 PowerSupply . . . . . . . . . . . . . . . . . . . . . . . . 70
5.2.6 EnclosureDesignandSystemIntegration . . . . . . . . . 73
6 SensorDesign 77
6.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
6.2 AmplifierDesign . . . . . . . . . . . . . . . . . . . . . . . . . . 79
6.3 ElectrodeDesign . . . . . . . . . . . . . . . . . . . . . . . . . . 83
6.4 HighImpedanceBiasing . . . . . . . . . . . . . . . . . . . . . . 87
6.5 InputCapacitanceNeutralisation . . . . . . . . . . . . . . . . . . 106
6.6 Integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
7 SensorEvaluation 117
7.1 PCBLeakage . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
7.2 FrequencyResponse . . . . . . . . . . . . . . . . . . . . . . . . 117
7.3 GainvsDistance . . . . . . . . . . . . . . . . . . . . . . . . . . 120
7.4 InputImpedanceEstimation . . . . . . . . . . . . . . . . . . . . 122
7.5 SettlingTime . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
7.6 NoiseMeasurement . . . . . . . . . . . . . . . . . . . . . . . . . 125
7.7 ECGRecording . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
8 ConclusionandRecommendationsforExtension 129
8.1 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
8.2 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
8.3 Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
8.4 Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . 132
CONTENTS vii
A SpiceSimulationforHighImpedanceCircuits 135
B SensorCalibration 137
B.1 LaboratoryEquipment . . . . . . . . . . . . . . . . . . . . . . . 137
B.2 Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
B.3 Notesforfuturecalibrators . . . . . . . . . . . . . . . . . . . . . 140
C CircuitDiagramsandPCBlayouts 141
viii CONTENTS