Small portable complex dielectric permittivity sensor-based system based on Wheatstone bridge
Technical field
The invention belongs to microwave and millimeter wave circuit and sensor technical field, propose it is a kind of based on Wheatstone bridge just
Formula sensor-based system is taken to measure complex dielectric permittivity.
Background technique
It is considered as a kind of up-and-coming tool about micro-wave dielectric spectrum in modern sensor-based system, and extensively
Applied to agricultural, food, auto industry and field of biomedicine.This method depends on measurement and Jie of complex dielectric permittivity
Matter complex dielectric permittivity with frequency variation.For example, the variation of vegetables and fruit complex dielectric permittivity is estimated that water in agricultural
With the variation of inorganic content;In automobile industry, the measurement of dielectric constant is preferred oil quality detection method;In biology
In terms of medicine, it is widely used in blood sugar test and internal cancer detection and assessment.Although dielectric constant measurement system is in the modern times
Huge potentiality are shown in life, but since traditional measurement mode tends to rely on large-scale experiment equipment such as vector network point
Analyzer.This restrict this method in outdoor scene, is remotely located the application of measurement etc., and cause the valuableness of cost.
On the other hand, in order to really develop potentiality of the microwave permittivity sensor in practical application, for sensor
Miniaturization Design with measuring system is essential.In addition, the sensor of miniaturization can promote new application, such as with
In in microwave frequency band realization dielectric constant real-time measurement and the design of visual two-dimensional sensor array.In order to meet it is such at
As systematic difference demand, it is necessary to focus in the size for how reducing sensor and its circuit for signal conditioning, need simultaneously
Data and the higher characteristic of resolution ratio can quickly be read by having.
In addition, dielectric constant is often a real number, i.e. ε in naturer *=εr′-jεr", wherein real part εr' generation
The table storage of energy, and imaginary part εr" then characterize the loss of material.Existing major part measurement method is often only focused in reality
The measurement in portion, and the imaginary part information of dielectric permittivity can not be obtained.Such as the resonant cavity perturbation method industrially most often utilized, lead to
The variation for crossing measuring medium loaded cavity front and back frequency is just unable to measure out imaginary values to obtain dielectric constant.
Therefore, above two aspect, which becomes, restricts an important factor for dielectric sensors are further applied.How it is succinct efficiently
It extracts and conversion sensor output signal, research and development is used for the miniaturization sensor-based system of dielectric constant measurement, and is how same
When extract real part of permittivity and imaginary part information, have become a research hotspot of academia and industry.
Summary of the invention
In view of this, the present invention is to propose a kind of small portable in order to overcome above-mentioned existing difficulty and answer dielectric
Constant sensing system.
The sensing system is mainly made of near field patch sensor, balanced Wheatstone bridge and lower frequency changer circuit.Wherein
Patch sensor is used to place testing medium, and Wheatstone bridge is used to measure the impedance of patch sensor.
The Wheatstone bridge and lower frequency changer circuit is well-known technique.
Near field patch sensor include the silicon base layer set gradually from bottom to up, metal ground layer, oxide layer, metal patch with
And passivation layer;Oxide layer upper surface is provided with a groove, metal patch is placed in the groove, and there are passivation layers above metal patch
Opening;
When detection, testing medium (MUT) is placed on sensor upper surface, and metal patch is completely covered, at this time Jie to be measured
There are passivation layer opening, that is, air layers between matter and metal patch, are equivalent to a fixed capacity;
Working principle is as follows: near field patch sensor is equivalent to a fixed capacity, when there is testing medium (MUT) to put
When going, resultant admittance can change, and the conductance and capacitor of the admittance are related with dielectric constant.
The near field patch sensor is replaced to the capacitor of one of branch road of Wheatstone bridge, which is believed by radio frequency
Number driving, by the available patch sensor impedance real part imaginary part information of the amplitude and phase of electric bridge differential output signal, most
Intermediate-freuqncy signal is exported by lower frequency changer circuit afterwards, extracts dielectric constant size by ADC and Digital Signal Processing.
The beneficial effects of the present invention are:
The present invention avoids the use of vector network analyzer, greatly reduces testing cost;It is normal dielectric can be measured simultaneously
Several real and imaginary parts.
Detailed description of the invention
Fig. 1 is patch sensor 3D schematic diagram in the present invention.
Fig. 2 is patch sensor technique cross-sectional view in the present invention.
Fig. 3 be in the present invention under 1GHz frequency total conductance of patch sensor and capacitor with medium real part imaginary part change
Change.
Fig. 4 is Wheatstone bridge basic principle block diagram in the present invention.
Fig. 5 is that bridge output signal is downconverted to intermediate frequency schematic circuit diagram in the present invention.
Fig. 6 is that balanced type difference electric bridge implements circuit diagram in the present invention.
In figure: 1.MUT;2. passivation layer;3. oxide layer;4. metal;5. silicon substrate;6. passivation layer opening;7. metal patch.
Specific embodiment
Following is a specific embodiment of the present invention in conjunction with the accompanying drawings, technical scheme of the present invention will be further described,
However, the present invention is not limited to these examples.
In view of the deficiencies in the prior art, it is found by the applicant that avoid can be with to the measurement of medium sensor frequency response
Avoid the dependence of the large-scale experiments equipment such as vector network analyzer.Therefore applicant is conceived to the detection to sensor admittance.Fig. 1
It is shown patch sensor 3D schematic diagram, since single metal patch electric field line is deeper, near-field field strength is bigger, therefore can be used as
One good sensing element.The near field patch sensor is designed using typical CMOS technology, technique cross-sectional view such as Fig. 2
It is shown.Lowest level is silicon base layer (5), connects metal ground layer (4) above silicon base layer, connects oxide skin(coating) (3) above metal ground layer,
Oxide layer connects passivation layer (2), and oxide layer digs groove insertion metal patch (7), and above metal patch
Passivation layer remove to form passivation layer opening (6).Passivation layer is placed above testing medium (MUT) (1) and metal is completely covered
Patch.Wherein metal layer material is copper, and oxide layer materials are silica, and silicon substrate layer material is silicon.
Above-mentioned near field patch sensor equivalent circuit is as shown in Fig. 2, patch sensor is equivalent to a fixed capacity C0, C0
One end ground connection, the other end are denoted as point P, the admittance Y of testing mediumMUTIt is equivalent to a capacitor CMUTWith a conductance GMUTParallel connection,
Its latter end ground connection in parallel, one end tie point P.
When not having testing medium placement, the admittance of node P is exactly that metal ground layer and metal patch layer are formed by fixation
Capacitor C0, when there is MUT to be put into, the admittance of P point will be with the variation of loaded medium and change, due to the reality of complex dielectric permittivity
Portion and imaginary part represent the energy storage and loss of medium respectively, as shown in Fig. 2, therefore dielectric layer can be equivalent to a capacitor and electricity
The parallel connection of resistance, admittance YMUT≈GMUT(ε″)+jωCMUT(ε '), then the resultant admittance of node P is YP=j ω C0+YMUT。
For the relationship of quantitative analysis P point admittance and dielectric constant, 100*100 a μm based on 40nm technique2Size
Patch sensor carries out EM Electromagnetic Simulation.Fig. 3 (1), Fig. 3 (2) show capacitor at 1 GHz and conductance with dielectric constant reality
The variation in portion and imaginary part.As seen from the figure, ε ' and capacitor, ε " and conductance are approximately a kind of linear relationship, that is, are had
YP(ε ', ε ", ω) ≈ αi·ω·ε″+jω·(C0+αrε′) (1)
Wherein αiAnd αrFor real parameter.Table one utilizes the parameters value after least square method fitting after summarizing emulation
Parameter |
Numerical value |
C0 |
82.56fF |
αr |
2.754fF-1 |
αi |
17.5μS-1·GHz-1 |
Since the capacitor and conductance of P point are also respectively by ε " and ε ' is influenced.Therefore applicant fits from EM emulation
A kind of rational function model is calibrated.The model formation is as follows
Wherein a is a scale parameter, is equal to patch sensor size, αnpAnd βmqIt is the reality of N*P and M*Q rank matrix
Model parameter.In order to make the model parameter meet EM emulation in admittance with ε ' and ε " variation, applicants have discovered that working as N=P=Q=
When M=1, error rate of the dielectric constant of calculating in 0.1-10GHz frequency range is within 1%.
By patch sensor, the measurement of complex dielectric permittivity is converted the measurement for admittance by the present invention, in the present invention
The admittance of patch sensor is measured using Wheatstone bridge, because the electric bridge provides one quantifies admittance measurement phase well
For a benchmark admittance, the C in the patch sensor0It is benchmark admittance.The electric bridge basic schematic diagram as shown in figure 4, its
It is mainly made of the branch that a driving amplifier and four groups of shunt capacitance conductances form, the input signal of electric bridge is vi, connection
V is exported after driving amplifierin,vinIt is divided into one end that two-way is connected to the first branch and third branch, the other end of the first branch
One end of second branch is connected, tie point is denoted as node A, voltage vB, O+, the other end of second branch be connected to ground, third
The other end of branch connects one end of the 4th branch, and tie point is denoted as node B, voltage vB, O-.The other end of 4th branch
It is connected to ground.Wherein this three branch admittances of the first branch, second branch and third branch are benchmark admittance Y1,Y1For fixation
Conductance G1With fixed capacity C1Parallel connection.The initial admittance of 4th branch is Y1, the patch of loaded medium is sensed on this basis
Device is connected into the 4th branch, then the 4th branch admittance becomes benchmark admittance Y1With load admittance YLThe sum of., the electric bridge is by a frequency
Rate is that the signal of ω is motivated through overdriven amplifier, A, the differential output signal and input signal v of B nodeinThe following institute of relationship
Show
Wherein YL=GL+jBL,Y1=G1+jB1, G is admittance real part conductance, and B is admittance imaginary part susceptance, it is assumed that
YL≠ 0, the inverse of formula (3) is taken, is had
Use GL, BL, G1And B1To replace YLAnd Y1, then have:
Wherein GLω=GL/|YL|2, BLω=BL/|YL|2, respectively weighted load conductance and susceptance value.By formula (5) and
(6) it is found that regardless of load admittance YLDeviate benchmark admittance Y1How much, the real and imaginary parts of electric bridge difference output are weighted loads
The linear combination of conductance and susceptance value.Based on this, pass through the linear relationship of electric bridge input and output, so that it may avoid higher order polynomial
Fitting and easily obtain load admittance value (real and imaginary parts), and then obtain dielectric constant information.But as shown in Figure 4
Circuit structure output end have a very big common-mode signal components, this to it is subsequent read chain circuit common-mode rejection ratio
(CMRR) strict requirements are proposed.In consideration of it, we use a kind of double balanced arrangement (double-balanced) differential electricals
Bridge design scheme solves the above problems.Double balanced arrangement difference electric bridges are made of two Wheatstone bridges, one of favour this
Energization bridge does not load admittance as reference bridge, another Wheatstone bridge has load admittance as measuring bridge, exports and is
The difference output of two electric bridges.
Meanwhile in order to facilitate subsequent digital signal processing, the present invention is used above-mentioned double balanced arrangement (double-
Balanced) output of difference electric bridge is down-converted to intermediate frequency and reconnects analog-digital converter (analog-to-digital
Converter, ADC) it is handled.As shown in figure 5, a double balanced mixer (double-balanced mixer) passes through
It connects electric bridge difference output and local oscillation signal and converts intermediate-freuqncy signal v for electric bridge radio frequency output signalIF+,vIF-.Since intermediate frequency is believed
Number frequency is lower, can be carried out facilitating processing by inexpensive ADC, be greatly lowered the cost of implementation of system in this way.In addition, this
Vibration LO signal is square-wave signal, since frequency mixer works in linear region, intermediate frequency difference output Δ vIFIt is poor that electric bridge has been fully retained
Divide the Δ v of outputB, oInformation, therefore Δ v can be passed throughIFLoad admittance information is obtained, and then obtains complex permittivity values.
Embodiment 1
The present invention measures the dielectric constant information of material using balanced type difference electric bridge shown in fig. 6.Patch sensor
Pass through a switch lcIt is connected in parallel to the third branch road of measuring bridge, as shown, the circuit is made of a pair of of difference electric bridge,
The benchmark admittance of branch is Cb.In order to adapt to the variation of extensive capacitive load, and to electric bridge behavior under various non-equilibrium states into
Row experimental study, C in implementation processbSpecifically by 8 switchable capacitors C3It composes in parallel, each capacitor C3Pass through series connection one
The realization of 10- μm/40-nm cmos switch freely opens and closes, and thus provides 8 kinds of benchmark admittance values, wherein C3Size be
100fF.In addition, resistance of connecting has been distinguished among electric bridge intermediate node C, D, C ' and D ' and ground, to ensure switch bias
When have a suitable DC channel.By formula (5) and (6) it is found that in order to measure the capacitor of load admittance and electric conductivity value, need
Obtain the amplitude and phase of electric bridge output.Especially, it should be noted that the consistency in order to guarantee phase measurement, it is also necessary to survey
Measure a fixed phase.This be in order to ensure the relative phase of bridge output change, only by patch load variation and
It is caused.For the relative amplitude and phase change of the output quantity of measurement dielectric material load front and back, the present invention is used in electric bridge
One reference capacitance C of parallel connection at load admittance branch, that is, node CfWith the patch sensor, the capacitor and sensor pass through
Switch lcRealize switching gating.CfOutput phase when gating is fixed phase.And when sensor gating, testing medium draws
Membership generates additional a phase offset and amplitude offset.It therefore, can be by the way that repeatedly switching is opened in a continued time domain
It closes, for obtaining the continuous measurement tracking of two output comprising electric bridge in the case where both load connection.By to acquisition
To signal carry out lower conversion and mould conversion process, and by synchronous with control switch lc, two independent can be exported in number
It is isolated in word domain.Then the Fast Fourier Transform (FFT) (FFT) of two output is calculated, and it is split, one can be obtained
The relative phase difference and Amplitude Ration of cause, and then obtain the real and imaginary parts of load admittance.
The above description of the embodiment is only used to help understand the method for the present invention and its core ideas.It should be pointed out that pair
For those skilled in the art, without departing from the principle of the present invention, the present invention can also be carried out
Some improvements and modifications, these improvements and modifications also fall within the scope of protection of the claims of the present invention.To these embodiments
A variety of modifications are it will be apparent that General Principle defined herein can be for those skilled in the art
It is realized in other embodiments in the case where not departing from the spirit or scope of the present invention.Therefore, the present invention is not intended to be limited to
These embodiments shown in the application, and be to fit to consistent with principle disclosed in the present application and features of novelty widest
Range.