CN103364752A - Field calibration method of on-wafer load traction measurement system - Google Patents

Abstract

The invention discloses a field calibration method of an on-wafer load traction measurement system. The field calibration method comprises the steps of firstly, manufacturing a mismatching attenuation chip with the reflection coefficient within the range of 0.1-0.8 to be used as a transmitting standard component, then, using a vector network analyzer which is calibrated in an on-wafer mode for carrying out calibration measurement on the transmitting standard component, obtaining a standard conversion gain GT(S) of the transmitting standard component and the calibration uncertainty of the transmitting standard component, then, using the transmitting standard component after the calibration for carrying out conversion gain parameter calibration on the on-wafer load traction measurement system, and completing the working of traceability of a quantity value of the on-wafer load traction measurement system. According to the calibration technology, the comprehensive and actual calibration can be carried out on the performance index of the on-wafer load traction measurement system, the calibration deviation delta GT is given, the measurement uncertainty of the on-wafer load traction measurement system can be quantitatively given, help is brought to the unification of the quantity value of the on-wafer load traction measurement system, and measurement technique supporting is provided for development and production of a power single chip circuit.

Description

A kind of field calibration method in sheet load traction measuring system

Technical field

The microwave/millimeter wave fields of measurement is about obtaining the accurately technology of large signal parameters performance of device by change source or loaded impedance.

Background technology

For linear unit, can extrapolate performance under any load by S parameter under the small-signal, the load traction method is not necessary.But the microwave power transistor output power is large, is generally operational under the large-scale condition, shows very strong nonlinear characteristic.Therefore, the traditional designing requirement that can't satisfy microwave power transistor under the large-signal condition based on the small-signal method for designing of linear theory.Various power amplifiers are widely used in the various electronics, and the design of power amplifier actual be exactly the matching network of developing between the power device, study which type of network and can obtain high output power, high efficient and needed gain, this output power, benefit and the gain that will understand device changes with loaded impedance and how to change, and load traction measuring system just can provide these information.The deviser compromises and selects the optimum matching impedance, designs the perfect match network, gives full play to the ability of device, obtains high output power, efficient and required gain.

Load traction measuring system is very complicated, particularly in the sheet measuring system, its overall performance index does not have effective metering measure, because the restriction of the factors such as the product of introduction different vendor and personnel, condition, the load Traction Parameters measurement result difference that constituent parts provides is very large, and causing sometimes two cover similar systems to test the curve of output power, gain, efficient under same impedance also can be inconsistent.For the value of realizing measuring system accurately and reliably, ensure the consistance that the load Traction Parameters is measured, provide strong measurement technology guarantee for improving the key electronic device test analysis and designing and developing ability, therefore, extremely be necessary to carry out the calibration operation in sheet load traction measuring system.

Summary of the invention

The invention provides a kind of field calibration method in sheet load traction measuring system, the method is for unified load traction measuring system parameter value, provides practical load Traction Parameters magnitude tracing approach to develop.

In order to solve the problems of the technologies described above, the technical solution used in the present invention is: a kind of field calibration method in sheet load traction measuring system comprises the steps:

The first step is developed the different mismatch attenuation monolithic of a series of reflection coefficients as Transfer Standards spare, and the coverage of the reflection coefficient of described Transfer Standards spare is: 0.1-0.8;

Second step utilizes vector network analyzer and microwave probe platform to form at the sheet vector network analyzer as robot scaling equipment, measures the S parameter of each Transfer Standards spare with robot scaling equipment, and described S parameter is S ₁₁, S ₁₂, S ₂₁, S ₂₂

In the 3rd step, by the source of the robot scaling equipment of setting, the S parameter that the load end reflection coefficient records in conjunction with previous step, calculate the standard handovers gain G of each Transfer Standards spare with reference to formula (1) _T(S), and with its calibration standard value as corresponding Transfer Standards spare conversion gain;

$G_T\left(S\right)=\frac{(1-{\left|{\mathrm{\Γ}}_S\right|}^2){\left|S_{21}\right|}^2(1-{\left|{\mathrm{\Γ}}_L\right|}^2)}{{|(1-S_{11}{\mathrm{\Γ}}_S)(1-S_{22}{\mathrm{\Γ}}_L)-S_{12}{S}_{21}{\mathrm{\Γ}}_S{\mathrm{\Γ}}_L|}^2}...\left(1\right)$

Wherein, Γ _SBe the reflection coefficient of looking to the signal source from Transfer Standards spare input end; Γ _LBe the reflection coefficient of looking to the output load end from Transfer Standards spare output terminal;

In the 4th step, calculate each Transfer Standards spare S parameter based on the uncertainty of measurement of robot scaling equipment;

In the 5th step, obtain because the standard handovers gain G that the indeterminacy of S parameter is introduced by partial differential equation _T(S) uncertainty of measurement, the uncertainty of measurement of again measuring repeatability being introduced be the synthetic standard handovers gain G that obtains with it _T(S) calibration uncertainty;

In the 6th step, be used in the measurement conversion gain G that sheet load traction measuring system is measured each Transfer Standards spare _T, obtain the uncertainty of measurement introduced in sheet load traction measuring system measuring repeatability, and the G that itself and above-mentioned steps are obtained _T(S) calibration uncertainty combination obtains the uncertainty of measurement in sheet load traction measuring system, is implemented in the calibration of sheet load traction measuring system.

The way of realization of Transfer Standards spare is in the above-mentioned first step:

In the design: adopt good stability, distribution parameter is little and the form of the balanced type ∏ type resistor network that suitable domain is arranged;

On the technique: described Transfer Standards spare utilizes sputtering technology take the gaas wafer sheet as substrate, and nickel-chrome is made the alternating-current resistance network at substrate, and the alternating-current resistance network is made the form of balanced type ∏ type resistor network, and covers silicon nitride film thereon; The resistor network that development is finished is made at the gallium arsenide disk with the form of microstrip line, and disk is done the via hole processing after being thinned to 100 μ m; Last Transfer Standards spare is preserved at disk.

The computing method in above-mentioned the 4th step are: measure the system spare error term of robot scaling equipment and stability error, noise error, and utilize these errors to calculate the uncertainty of measurement of S parameter.

Above-mentioned the 5th step Plays conversion gain G _TThe computing method of calibration uncertainty (S) are:

At first, be the form that mould and phase place combine, i.e. S with each S parameter decomposition ₁₁=r ₁₁∠ θ ₁₁, S ₂₁=r ₂₁∠ θ ₂₁, S ₁₂=r ₁₂∠ θ ₁₂, S ₂₂=r ₂₂∠ θ ₂₂, the standard handovers gain G _T(S) mould of each S parameter and phase place are carried out partial differential and obtain formula (2):

$d{G}_T\left(S\right)=\frac{\∂G_T\left(S\right)}{\∂r_{11}}{\mathrm{dr}}_{11}+\frac{\∂G_T\left(S\right)}{\∂{\mathrm{\θ}}_{11}}d{\mathrm{\θ}}_{11}+\frac{\∂G_T\left(S\right)}{\∂r_{21}}{\mathrm{dr}}_{21}+\frac{\∂G_T\left(S\right)}{\∂{\mathrm{\θ}}_{21}}d{\mathrm{\θ}}_{21}$

$+\frac{\∂G_T\left(S\right)}{\∂r_{12}}d{r}_{12}+\frac{\∂G_T\left(S\right)}{\∂{\mathrm{\θ}}_{12}}d{\mathrm{\θ}}_{12}+\frac{\∂G_T\left(S\right)}{\∂r_{22}}d{r}_{22}+\frac{\∂G_T\left(S\right)}{\∂{\mathrm{\θ}}_{22}}d{\mathrm{\θ}}_{22}$ ..................（9）

Wherein, dr ₁₁, dr ₂₁, dr ₁₂, dr ₂₂Be the uncertainty of the mould of four S parameters, d θ ₁₁, d θ ₂₁, d θ ₁₂, d θ ₂₂It is the phase place uncertainty of four S parameters;

Then, will be decomposed into the uncertainty of measurement of mould and the uncertainty of measurement of phase place by the 4th uncertainty of measurement that goes on foot the S parameter that obtains;

Then, according to the uncertainty of measurement of mould and the uncertainty of measurement of phase place, calculate the standard handovers gain G with reference to formula (9) _T(S) uncertainty of measurement;

At last, calculate robot scaling equipment because the uncertainty of measurement that measuring repeatability is introduced, with itself and G _T(S) the synthetic standard handovers gain G that obtains of uncertainty of measurement _T(S) calibration uncertainty.

Mentality of designing of the present invention is as follows:

Figure 1 shows that the structural drawing in sheet load traction measuring system.Load traction measuring system has the ability of the reflection coefficient of setting source/load end, usually the uncertainty of measurement of load traction measuring system be mainly derived from source/load end reflection coefficient whether accurately and the output terminal power measurement whether accurate, and such uncertainty of measurement can not directly be measured, and therefore just expects using the mode of conversion indirectly to measure uncertainty.The physical quantity of indirectly measuring should be relevant with the reflection coefficient of source/load end, and other parameters are not introduced uncertainty of measurement in load traction measuring system as far as possible in this physical quantity, and the uncertainty of measurement that embodies of this physical quantity could be converted to the uncertainty of measurement that measuring system is drawn in load more accurately like this.And conversion gain G _T(S) just in time meet this requirement.Conversion gain G _T(S) be with the parameter (four S parameters) of Transfer Standards spare itself and be calibrated the source impedance Γ of measuring system _sWith loaded impedance Γ _LRelevant parameters all, the uncertainty of measurement of the load traction measuring system that the S parameter is introduced is very little, simultaneously conversion gain G _TAlso comprised the sign to output terminal power measurement result, therefore chosen this parameter and can the rational evaluation load draw the measurement performance of measuring system as the parameter of tracing to the source that is calibrated system, calibration result is more credible.

For unified load traction measuring system parameter value, practical load Traction Parameters magnitude tracing approach is provided, calibration steps of the present invention be mismatch attenuation monolithic by the different reflection coefficients that design and produce stable performance (cover 0.1 to 0.8) as Transfer Standards spare, be used in vector network analyzer that sheet calibrated as the standard handovers gain G of robot scaling equipment to Transfer Standards spare _T(S) calibrate, obtain the standard handovers gain G _T(S) and the calibration uncertainty; Be used in again sheet load traction measuring system to the conversion gain G of the Transfer Standards spare after calibrating _TCarry out in-site measurement, with the standard handovers gain G _T(S) and measure conversion gain G _TCompare and obtain △ G _T, as modified value, finish in the work of tracing to the source of sheet load traction measuring system value.Fig. 2 is calibrating principle synoptic diagram of the present invention.

Discuss in view of top analysis to calibration steps, the technical progress of adopting technique scheme to obtain is: calibrate by development Transfer Standards spare and to it, can make things convenient for, finish accurately the field calibration in sheet load traction measuring system; This collimation technique can to carrying out comprehensively in the performance index of sheet load trailer system, calibrating really, provide modified value △ G _TAnd the calibration uncertainty that is given in sheet load traction measuring system that can be quantitative, realize the measurement result value of the sheet load traction measuring system that do not coexist in the engineering application accurately, reliably, the consistance of carrying out designs in sheet load traction measuring system after calibrating is significantly improved; The small volume of Transfer Standards spare among the present invention, the probe spacing of employing standard is applicable to various existing measuring systems, has improved the universality of this method.

Description of drawings

Fig. 1 is the structural drawing in sheet load traction measuring system;

Fig. 2 is the calibrating principle synoptic diagram;

Fig. 3 is Transfer Standards spare project organization exemplary plot;

Fig. 4 is the structural representation of robot scaling equipment.

Embodiment

Below in conjunction with concrete embodiment the present invention is carried out more detailed explanation and explanation.

A kind of field calibration method in sheet load traction measuring system specifically comprises the steps:

The first step is developed the mismatch attenuation monolithic of a series of different reflection coefficients as Transfer Standards spare, and the reflection coefficient coverage of described Transfer Standards spare is: 0.1-0.8.

The present invention calibrates the Transfer Standards spare of development by building robot scaling equipment, be unified in the value of sheet load trailer system, provide practical in sheet load trailer system magnitude tracing approach, so the development of Transfer Standards spare and calibration work are the important steps that realizes magnitude tracing.

Stable for guaranteeing in broad frequency range, to have, repeatably, little gain characteristic affected by environment, the accuracy that improves this method, needing to adopt passive device workpiece be Transfer Standards spare.Set about from the sputtering technology of Transfer Standards spare and the homogeneity aspect of resistance alloys, design and produce the Transfer Standards spare that in broad frequency range, meets performance requirement.

In design: the form with GaAs material microstrip line is made, and adopts the form of balanced type ∏ type resistor network, can improve the stability of Transfer Standards spare; In order to reduce distribution parameter and to be fit to the spacing requirement of linking probe, the project organization form of Transfer Standards spare as shown in Figure 3: 602 Ohmages of input end and output terminal, the form that can be designed to respectively two 1204 ohm parallel connection reduces distribution parameter.

On technique: standard component is take the gaas wafer sheet as substrate, utilize sputtering technology, nickel-chrome is made the alternating-current resistance network at substrate, the alternating-current resistance network is made the form of balanced type ∏ type resistor network, being illustrated in figure 3 as damping capacity is the structural representation of 6dB, reflection coefficient 0.6, covers silicon nitride film thereon; The resistor network that development is finished is made at the gallium arsenide disk with the form of microstrip line, and disk is done the via hole processing after being thinned to 100 μ m; Last Transfer Standards spare is just preserved at disk.

The technical feature of a series of Transfer Standards spares that this invention is made is as follows:

A, frequency range: 2GHz-18GHz;

B, reflection coefficient are respectively 0.1,0.2, and 0.3,0.4,0.5,0.6,0.7,0.8;

Reflection coefficient precision＜± 0.05

C, when damping capacity is 3dB, reflection coefficient is respectively 0.1,0.2,0.3;

When damping capacity was 6dB, reflection coefficient was respectively 0.1,0.2, and 0.3,0.4,0.5;

When damping capacity is 15dB, reflection coefficient difference 0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8;

Attenuation accuracy＜0.1dB

Technological parameter: the GSG spacing is 150 μ m, and pressure point is 50 μ m * 50 μ m.

Need to prove: the calibration precision of Transfer Standards spare depends on the accuracy of robot scaling equipment and the stability of Transfer Standards spare, does not have direct relation with the design objective (such as reflection coefficient＜0.05) of Transfer Standards spare.

Second step utilizes vector network analyzer and microwave probe platform to form at the sheet vector network analyzer as robot scaling equipment, as shown in Figure 4.Measure the S parameter of each Transfer Standards spare with robot scaling equipment, described S parameter is S ₁₁, S ₁₂, S ₂₁, S ₂₂

On the microwave probe platform, vector network analyzer is TRL or LRM calibrates at sheet, to reduce the calibration uncertainty of vector network analyzer, end face can be calibrated to probe location at the sheet vector network analyzer like this, realized going embedding calibration, as shown in Figure 4.Use this robot scaling equipment to measure the S parameter of each Transfer Standards spare, described S parameter comprises S ₁₁, S ₁₂, S ₂₁, S ₂₂

In the 3rd step, by the source of the robot scaling equipment of setting, the S parameter that the load end reflection coefficient records in conjunction with previous step, calculate the standard handovers gain G of each Transfer Standards spare with reference to formula (1) _T(S), and with its calibration standard value as corresponding Transfer Standards spare conversion gain;

$G_T\left(S\right)=\frac{(1-{\left|{\mathrm{\Γ}}_S\right|}^2){\left|S_{21}\right|}^2(1-{\left|{\mathrm{\Γ}}_L\right|}^2)}{{|(1-S_{11}{\mathrm{\Γ}}_S)(1-S_{22}{\mathrm{\Γ}}_L)-S_{12}{S}_{21}{\mathrm{\Γ}}_S{\mathrm{\Γ}}_L|}^2}...\left(1\right)$

Wherein, Γ _SBe the reflection coefficient of looking to the signal source from Transfer Standards spare input end; Γ _LBe the reflection coefficient of looking to the output load end from Transfer Standards spare output terminal.

Formula (1) can obtain formula (2) through conversion:

$G_T\left(S\right)=\frac{(1-{\left|{\mathrm{\Γ}}_S\right|}^2){\left|S_{21}\right|}^2(1-{\left|{\mathrm{\Γ}}_L\right|}^2)}{{|1-{\mathrm{\Γ}}_{\mathrm{in}}{\mathrm{\Γ}}_S|}^2{|1-S_{22}{\mathrm{\Γ}}_L|}^2}...\left(2\right)$

Γ wherein _InBe the reflection coefficient of looking from Transfer Standards spare input end to load end;

${\mathrm{\Γ}}_{\mathrm{in}}=S_{11}+\frac{S_{21}{S}_{12}{\mathrm{\Γ}}_L}{1-S_{22}{\mathrm{\Γ}}_L}...\left(3\right)$

Conversion gain is defined as: the power that is sent to load and ratio from the available power in source, definition is worked as Γ as can be known thus _SWith Γ _InDuring conjugate impedance match

Available power from the source is the peak power that can be sent to network.

Have the ability of the reflection coefficient of setting source/load end in sheet load traction measuring system, therefore according to the power that is sent to load in the conversion gain definition, but Set arbitrarily Γ _L, for passive Transfer Standards spare, when

The time can obtain conversion gain G by formula (2) _T(S) variation formula (4), it is actual to be exactly the power gain of device.

$G_T\left(S\right)=\frac{{\left|S_{21}\right|}^2(1-{\left|{\mathrm{\Γ}}_L\right|}^2)}{(1-{\left|{\mathrm{\Γ}}_{\mathrm{in}}\right|}^2){|1-S_{22}{\mathrm{\Γ}}_L|}^2}...\left(4\right)$

Therefore actual in when sheet load traction measuring system is carried out calibration measurement, for the loaded impedance Γ of Set arbitrarily _L, can be reached by formula (3) Calculate fast corresponding source impedance Γ _s, and do not need to find corresponding source impedance Γ by the way of traction _s, and can calculated in advance go out the respective standard conversion gain G of Transfer Standards spare _T(S).This is very convenient to carrying out field calibration and providing the uncertainty of calibration result.

In the 4th step, calculate each Transfer Standards spare S parameter based on the uncertainty of measurement of robot scaling equipment.

By top analysis as seen: the calibration precision of Transfer Standards spare is mainly determined by the measuring accuracy of robot scaling equipment.For this reason, need record the system spare error term (skin tracking error, source matching error, directional error) of robot scaling equipment and stability error, noise error etc., the synthetic uncertainty of measurement that calculates robot scaling equipment.This method with robot scaling equipment to the uncertainty of measurement of the S parameter main source as the uncertainty of measurement of Transfer Standards spare.The below take Transfer Standards spare in the 10GHz frequency as example, calculate robot scaling equipment to the uncertainty of measurement of S parameter.

With S ₁₁Be example, the computing formula of uncertainty is as follows:

$\mathrm{\&Delta;}{S}_{11\left(\mathrm{mag}\right)}=\sqrt{{\mathrm{Systematic}}^2+{\mathrm{<figure-callout\; id="2"\; label="Stability"\; filenames="130719090611.png"\; state="\{\{state\}\}">Stability</figure-callout>}}^2+{\mathrm{<figure-callout\; id="2"\; label="Noise"\; filenames="130719090611.png"\; state="\{\{state\}\}">Noise</figure-callout>}}^2}...\left(5\right)$

Δ S herein _{11 (mag)}Be S ₁₁Uncertainty of measurement, system spare error Systematic wherein ², the uncertainty Stability that introduces of stability and noise ²And Noise ²All can vow that the net error produces formula and obtains according to the PNA of Agilent company.

The system spare error is as the formula (6):

${\mathrm{Systematic}}^2={\mathrm{<figure-callout\; id="2"\; label="E\; DF"\; filenames="130719090611.png"\; state="\{\{state\}\}">E</figure-callout>}}_{\mathrm{DF}}^{}+E_{\mathrm{RF}}^2{S}_{11}^2+E_{\mathrm{XF}}^2{S}_{11}^4+S_{21}^2{S}_{12}^2({\mathrm{<figure-callout\; id="2"\; label="E\; LF"\; filenames="130719090611.png"\; state="\{\{state\}\}">E</figure-callout>}}_{\mathrm{LF}}^{}+4E_{\mathrm{SF}}^2{E}_{\mathrm{LF}}^2{S}_{11}^2+E_{\mathrm{LF}}^4{S}_{22}^2)+A_M^2{S}_{11}^2...\left(6\right)$

Every system spare error can measure respectively according to the generation principle of ten binomial errors, the results are shown in Table shown in 1:

Table 1 a vector network analyzer system spare error term measurement result

Content measurement	Measured value (dB)
		The effective source of forward coupling E SF	-32
Oppositely effectively E is mated in the source SR	-37
		Forward effective directivity E DF	-40
Reverse effective directivity E DR	-43
		Forward service load coupling E LF	-44
Oppositely E is mated in service load LR	-45
		The forward E that crosstalks XF	-61
E oppositely crosstalks XR	-61
		E is followed the tracks of in the forward transmission TF	0.06
Reverse transfer is followed the tracks of E TR	0.05
		Righting reflex is followed the tracks of E RF	0.11
Back reflection is followed the tracks of E RR	0.04
		Amplitude dynamic accuracy A M	0.02
Phase place dynamic accuracy A P（°）	5.75

The uncertainty that stability and noise are introduced also can be obtained by measurement and according to formula (7), (8):

${\mathrm{<figure-callout\; id="2"\; label="Stability"\; filenames="130719090611.png"\; state="\{\{state\}\}">Stability</figure-callout>}}^2=\sqrt{{\mathrm{<figure-callout\; id="2"\; label="C"\; filenames="130719090611.png"\; state="\{\{state\}\}">C</figure-callout>}}^2+{\mathrm{<figure-callout\; id="2"\; label="R"\; filenames="130719090611.png"\; state="\{\{state\}\}">R</figure-callout>}}^2}...\left(7\right)$

${\mathrm{<figure-callout\; id="2"\; label="Noise"\; filenames="130719090611.png"\; state="\{\{state\}\}">Noise</figure-callout>}}^2={\left(N_T{S}_{11}\right)}^2+{\mathrm{<figure-callout\; id="2"\; label="N\; F"\; filenames="130719090611.png"\; state="\{\{state\}\}">N</figure-callout>}}_F^{}...\left(8\right)$

Wherein, ${\mathrm{<figure-callout\; id="2"\; label="C"\; filenames="130719090611.png"\; state="\{\{state\}\}">C</figure-callout>}}^2=C_{\mathrm{RM}1}^2(1+S_{11}^4)+{4C}_{\mathrm{TM}1}^2{S}_{11}^2+{\mathrm{<figure-callout\; id="2"\; label="C\; RM"\; filenames="130719090611.png"\; state="\{\{state\}\}">C</figure-callout>}}_{\mathrm{RM}}^2{S}_{21}^2{S}_{12}^2$

${\mathrm{<figure-callout\; id="2"\; label="R"\; filenames="130719090611.png"\; state="\{\{state\}\}">R</figure-callout>}}^2={(R_{R1}(1+S_{11}^2)+{2R}_{T1}{S}_{11})}^2+{\left(R_{R2}{S}_{21}{S}_{12}\right)}^2$

In the following formula except the S parameter, parameters C _RM, C _TMBe cable stability component; R _R1, R _T1Be the connection repeatability component of joint, N _TBe the trace noise; N _FFor making an uproar at the end.

S ₂₁, S ₁₂, S ₁₂Can vow that the net error produces formula and calculates according to above-mentioned steps according to the PNA of Agilent company.

In the 5th step, obtain because the standard handovers gain G that the indeterminacy of S parameter is introduced by partial differential equation _T(S) uncertainty of measurement, uncertainty of measurement and the standard handovers gain G again measuring repeatability introduced _T(S) the synthetic Transfer Standards spare standard handovers gain G that obtains of uncertainty of measurement _T(S) calibration uncertainty.

Inevitably can introduce uncertainty when utilizing robot scaling equipment to measure the S parameter, and the uncertainty of S parameter can be married again the standard handovers gain G _T(S) become G on _T(S) based on the uncertainty of measurement of robot scaling equipment.

The standard handovers gain G _TThe computing method of uncertainty of measurement (S) are:

At first, be the form that mould and phase place combine, i.e. S with each S parameter decomposition ₁₁=r ₁₁∠ θ ₁₁, S ₂₁=r ₂₁∠ θ ₂₁, S ₁₂=r ₁₂∠ θ ₁₂, S ₂₂=r ₂₂∠ θ ₂₂, the standard handovers gain G _T(S) mould of S parameter and phase place are carried out partial differential and obtain formula (9):

$d{G}_T\left(S\right)=\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;\left|S_{11}\right|}d\left|S_{11}\right|+\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;{\mathrm{\&theta;}}_{11}}d{\mathrm{\&theta;}}_{11}+\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;\left|S_{21}\right|}d\left|S_{21}\right|+\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;{\mathrm{\&theta;}}_{21}}d{\mathrm{\&theta;}}_{21}$

$+\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;\left|S_{12}\right|}d\left|S_{12}\right|+\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;{\mathrm{\&theta;}}_{12}}d{\mathrm{\&theta;}}_{12}+\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;\left|S_{22}\right|}d\left|S_{22}\right|+\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;{\mathrm{\&theta;}}_{22}}d{\mathrm{\&theta;}}_{22}$

$=\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;r_{11}}{\mathrm{dr}}_{11}+\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;{\mathrm{\&theta;}}_{11}}d{\mathrm{\&theta;}}_{11}+\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;r_{21}}d{r}_{21}+\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;{\mathrm{\&theta;}}_{21}}d{\mathrm{\&theta;}}_{21}...\left(9\right)$

$+\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;r_{12}}d{r}_{12}+\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;{\mathrm{\&theta;}}_{12}}d{\mathrm{\&theta;}}_{12}+\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;r_{22}}d{r}_{22}+\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;{\mathrm{\&theta;}}_{22}}d{\mathrm{\&theta;}}_{22}$

Wherein, dr ₁₁, dr ₂₁, dr ₁₂, dr ₂₂Be the uncertainty of the mould of four S parameters, d θ ₁₁, d θ ₂₁, d θ ₁₂, d θ ₂₂It is the phase place uncertainty of four S parameters.

The uncertainty of measurement of the S parameter that then, above-mentioned four-step calculation is obtained is decomposed into the uncertainty of mould and the uncertainty of measurement of phase place.

Then, according to the uncertainty of measurement of mould and the uncertainty of measurement of phase place, calculate the standard handovers gain G with reference to formula (9) _T(S) uncertainty of measurement.

In the formula (9),

With

Calculation procedure as follows:

At first, formula (9) is converted to formula (10):

$G_T\left(S\right)=\frac{(1-{\left|{\mathrm{\&Gamma;}}_S\right|}^2){\left|S_{21}\right|}^2(1-{\left|{\mathrm{\&Gamma;}}_L\right|}^2)}{{|1-S_{22}{\mathrm{\&Gamma;}}_L-S_{12}{S}_{21}{\mathrm{\&Gamma;}}_S{\mathrm{\&Gamma;}}_L+S_{11}(S_{22}{\mathrm{\&Gamma;}}_S{\mathrm{\&Gamma;}}_L-{\mathrm{\&Gamma;}}_S)|}^2}...\left(10\right)$

Make Γ _S=r _S∠ θ _S, Γ _L=r _L∠ θ _L, σ=(1-Γ _S| ²) (1-| Γ _L| ²), R ₁₁=S ₂₂Γ _SΓ _L-Γ _S=r _R11∠ θ _R11, P ₁₁=1-S ₂₂Γ _L-S ₁₂S ₂₁Γ _SΓ _L,

Then formula (10) but simplified style (11):

$G_T\left(S\right)=\frac{\mathrm{\&sigma;}{r}_{21}^2}{{|P_{11}+S_{11}{R}_{11}|}^2}=\frac{\mathrm{\&sigma;}{r}_{21}^2}{{|\frac{P_{11}}{R_{11}}+S_{11}|}^2{\left|R_{11}\right|}^2}=\frac{\mathrm{\&sigma;}{r}_{21}^2}{{|Q_{11}+S_{11}|}^2{\left|R_{11}\right|}^2}...\left(11\right)$

$=\frac{\mathrm{\&sigma;}{r}_{21}^2}{r_{R11}^2[r_{11}^2+r_{Q11}^2-{2r}_{11}{r}_{Q11}\cos ({\mathrm{\&theta;}}_{11}-{\mathrm{\&theta;}}_{Q11})]}$

With

The example that is calculated as be introduced, with formula (11) to r ₁₁And θ ₁₁Carry out partial differential and obtain formula (12) and formula (13):

$\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;r_{11}}=-\frac{{2G}_T^2\left(S\right)}{\mathrm{\&sigma;}{r}_{21}^2}{r}_{R11}^2[r_{11}+r_{Q11}\cos ({\mathrm{\&theta;}}_{11}-{\mathrm{\&theta;}}_{Q11})]...\left(12\right)$

$\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;{\mathrm{\&theta;}}_{11}}=\frac{{2G}_T^2\left(S\right)}{{\mathrm{\&sigma;r}}_{21}^2}{r}_{R11}^2{r}_{11}{r}_{Q11}\sin ({\mathrm{\&theta;}}_{11}-{\mathrm{\&theta;}}_{Q11})...\left(13\right)$

In like manner can derive

With

These results are updated in the formula (9) again can obtain G _T(S) uncertainty of measurement u ₁

At last, also can not introduce uncertainty of measurement u because the robot scaling equipment measurement is repeated ₂, therefore calculate u ₂After with u ₁With u ₂The synthetic standard handovers gain G that obtains _T(S) calibration uncertainty u _C1u ₂Computing method be that those skilled in the art know, repeat no more here.

In the 6th step, be used in the conversion gain G that sheet load traction measuring system is measured each Transfer Standards spare _T, obtain the uncertainty of measurement introduced in sheet load traction measuring system measuring repeatability, and the G that itself and above-mentioned steps are obtained _T(S) calibration uncertainty combination obtains the uncertainty of measurement in sheet load traction measuring system, is implemented in the calibration of sheet load traction measuring system.

Measure the conversion gain G of each Transfer Standards spare in sheet load traction measuring system _TThe time, need the reflection coefficient of setting source/load end, at this moment, these reflection coefficients will be consistent with the numerical value set in the robot scaling equipment in above-mentioned the 3rd step, like this guarantee correctness of calibrating.

Will be at the uncertainty of measurement u of sheet load traction measuring system measuring repeatability introducing ₃The G that obtains with above-mentioned steps _T(S) calibration uncertainty u _C1The synthetic measurement uncertainty of measurement u that obtains in sheet load traction measuring system _c, be implemented in the calibration of sheet load traction measuring system.

The below is described in detail in the computing method of sheet load traction measuring system uncertainty of measurement with a concrete example.

With the Transfer Standards spare of damping capacity 15dB, the reflection coefficient 0.5 standard handovers gain G at the f=10GHz place _T(S) calibration uncertainty is that example is carried out the evaluation of uncertainty in measurement in sheet load traction measuring system.

The uncertainty of being measured the introducing of S parameter by robot scaling equipment is dG _T(S): u ₁=0.186dB;

Uncertainty of measurement u by the introducing of robot scaling equipment measuring repeatability ₂=0.050dB;

Both synthesize the standard handovers gain G that obtains this Transfer Standards spare _T(S) calibration uncertainty is:

$u_{c1}=\sqrt{{u_1}^2+{{\mathrm{<figure-callout\; id="2"\; label="u"\; filenames="130719090611.png"\; state="\{\{state\}\}">u</figure-callout>}}_2}^2}=0.193\mathrm{dB}$

With above-mentioned Transfer Standards spare at the 10GHz place, to the 0.8GHz-18GHz frequency range in sheet load traction measuring system (Tuner model: MT982BU) calibrate.Utilize this system to conversion gain G _TCarry out the result following (dB of unit) of 6 measurements:

-15.486，-15.491，-15.495,-19.485，-15.481，-15.485，

Can obtain the uncertainty u that this system is introduced by measuring repeatability by the above results ₃=0.035dB,

U then _C1And u ₃The synthetic uncertainty of measurement that obtains this system

The expanded uncertainty U=0.39dB(k=2 of this system);

This is the uncertainty of measurement net result in sheet load traction measuring system.

Following for result's of the present invention checking and application note.

Utilize the present invention to the 0.8GHz-18GHz frequency range in sheet load traction measuring system (Tuner model: MT982BU) try calibration.Carry out the parameter measurement at sheet S by the mismatch attenuation device that reflection coefficient is respectively 0.1,0.5,0.8 damping capacity 15dB, 10GHz frequency place calibration result is shown in Table 2:

Table 210GHz frequency is in sheet load traction measuring system calibration result

Do not provide uncertainty of measurement in sheet load traction measuring system, by the calibration of the present invention to system, just can access the uncertainty of measurement result.From the standard handovers gain G _T(s) and systematic survey conversion gain G _TDifference DELTA G _TCan find out with the numeric ratio of uncertainty: should be reliable at sheet load traction measuring system measurement result value.Δ G when large reflection coefficient (Γ=0.8) _TLarger, analyze reason and mainly contain following 3 points:

1) when sheet load traction measuring system was carried out self calibration, the S parameter of large reflection coefficient can produce larger uncertainty;

Also can produce larger uncertainty when 2) Transfer Standards spare carries out the S parameter calibration when large reflection coefficient;

3) this load trailer system is when sheet is measured monolithic integrated circuit, Γ=0.8th, and its limit is measured situation, and optimal match point can not accurately be calculated and find in system, can bring extra uncertainty.

In addition, (the Tuner model: MT984AU) the transmission part with Γ=0.5 is calibrated at the 10GHz place, relatively two to be enclosed within the measurement consistance that measuring system is drawn in the sheet load in sheet load traction measuring system to another 8GHz-50GHz also to utilize method of the present invention.The comparing data of two systems is shown in Table 3:

Table 3 uses same Transfer Standards spare to be enclosed within sheet load traction measuring system measurement data relatively to two

Parameter	0.8GHz-18GHz frequency range	The 8GHz-50GHz frequency range
			Γ S(mould/phase angle)	0.40∠10.26	0.55∠10.97
Γ L(mould/phase angle)	0.5∠-90	0.5∠-90
			The standard handovers gain G T(s)	-15.48dB	-15.46dB
Systematic survey conversion gain G T	-15.54dB	-15.74dB
			△G T	0.06dB	0.28dB

From data as can be known, should be in sheet load traction measuring system in the measurement gain results of 8GHz-50GHz frequency range than the measurement result of 0.8GHz-18GHz frequency range 0.22dB less than normal.As seen in order to make each load traction measuring system value accurately and reliably, can carry out corresponding correction to measuring system according to calibration result and make and respectively be calibrated system and have better consistance, realize the purpose of upwards tracing to the source.

By checking example and interpretation of result, this calibration steps is by carrying out the field calibration method research in sheet load traction measuring system, and development, demarcate the approach such as Transfer Standards spare and solved calibration problem in sheet load traction measuring system, unified helpful to the value that is implemented in sheet load traction measuring system, for the development of Power Monolithic Circuit, produce the measurement technology support is provided.

Claims (4)

1. the field calibration method in sheet load traction measuring system is characterized in that comprising the steps:

The first step is developed the different mismatch attenuation monolithic of a series of reflection coefficients as Transfer Standards spare, and the coverage of the reflection coefficient of described Transfer Standards spare is: 0.1-0.8;

Second step utilizes vector network analyzer and microwave probe platform to form at the sheet vector network analyzer as robot scaling equipment, measures the S parameter of each Transfer Standards spare with robot scaling equipment, and described S parameter is S ₁₁, S ₁₂, S ₂₁, S ₂₂

In the 3rd step, by the source of the robot scaling equipment of setting, the S parameter that the load end reflection coefficient records in conjunction with previous step, calculate the standard handovers gain G of each Transfer Standards spare with reference to formula (1) _T(S), and with its calibration standard value as corresponding Transfer Standards spare conversion gain;

$G_T\left(S\right)=\frac{(1-{\left|{\mathrm{\&Gamma;}}_S\right|}^2){\left|S_{21}\right|}^2(1-{\left|{\mathrm{\&Gamma;}}_L\right|}^2)}{{|(1-S_{11}{\mathrm{\&Gamma;}}_S)(1-S_{22}{\mathrm{\&Gamma;}}_L)-S_{12}{S}_{21}{\mathrm{\&Gamma;}}_S{\mathrm{\&Gamma;}}_L|}^2}...\left(1\right)$

Wherein, Γ _SBe the reflection coefficient of looking to the signal source from Transfer Standards spare input end; Γ _LBe the reflection coefficient of looking to the output load end from Transfer Standards spare output terminal;

In the 4th step, calculate each Transfer Standards spare S parameter based on the uncertainty of measurement of robot scaling equipment;

In the 5th step, obtain because the standard handovers gain G that the indeterminacy of S parameter is introduced by partial differential equation _T(S) uncertainty of measurement, the uncertainty of measurement of again measuring repeatability being introduced be the synthetic standard handovers gain G that obtains with it _T(S) calibration uncertainty;

In the 6th step, be used in the measurement conversion gain G that sheet load traction measuring system is measured each Transfer Standards spare _T, obtain the uncertainty of measurement introduced in sheet load traction measuring system measuring repeatability, and the G that itself and above-mentioned steps are obtained _T(S) calibration uncertainty combination obtains the uncertainty of measurement in sheet load traction measuring system, is implemented in the calibration of sheet load traction measuring system.

2. a kind of field calibration method in sheet load traction measuring system according to claim 1 is characterized in that the way of realization of Transfer Standards spare in the above-mentioned first step is:

In the design: adopt good stability, distribution parameter is little and the form of the balanced type ∏ type resistor network that suitable domain is arranged;

On the technique: described Transfer Standards spare utilizes sputtering technology take the gaas wafer sheet as substrate, and nickel-chrome is made the alternating-current resistance network at substrate, and the alternating-current resistance network is made the form of balanced type ∏ type resistor network, and covers silicon nitride film thereon; The resistor network that development is finished is made at the gallium arsenide disk with the form of microstrip line, and disk is done the via hole processing after being thinned to 100 μ m; Last Transfer Standards spare is preserved at disk.

3. a kind of field calibration method in sheet load traction measuring system according to claim 1, the computing method that it is characterized in that described the 4th step are: measure the system spare error term of robot scaling equipment and stability error, noise error, and utilize these errors to calculate the uncertainty of measurement of S parameter.

4. a kind of field calibration method in sheet load traction measuring system according to claim 1 is characterized in that above-mentioned the 5th step Plays conversion gain G _TThe computing method of calibration uncertainty (S) are:

At first, be the form that mould and phase place combine, i.e. S with each S parameter decomposition ₁₁=r ₁₁∠ θ ₁₁, S ₂₁=r ₂₁∠ θ ₂₁, S ₁₂=r ₁₂∠ θ ₁₂, S ₂₂=r ₂₂∠ θ ₂₂, the standard handovers gain G _T(S) mould of each S parameter and phase place are carried out partial differential and obtain formula (2):

$d{G}_T\left(S\right)=\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;r_{11}}{\mathrm{dr}}_{11}+\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;{\mathrm{\&theta;}}_{11}}d{\mathrm{\&theta;}}_{11}+\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;r_{21}}{\mathrm{dr}}_{21}+\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;{\mathrm{\&theta;}}_{21}}d{\mathrm{\&theta;}}_{21}$ ..................（9）

$+\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;r_{12}}d{r}_{12}+\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;{\mathrm{\&theta;}}_{12}}d{\mathrm{\&theta;}}_{12}+\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;r_{22}}{\mathrm{dr}}_{22}+\frac{\&PartialD;G_T\left(S\right)}{\&PartialD;{\mathrm{\&theta;}}_{22}}d{\mathrm{\&theta;}}_{22}$

Wherein, dr ₁₁, dr ₂₁, dr ₁₂, dr ₂₂Be the uncertainty of the mould of four S parameters, d θ ₁₁, d θ ₂₁, d θ ₁₂, d θ ₂₂It is the phase place uncertainty of four S parameters;

Then, will be decomposed into the uncertainty of measurement of mould and the uncertainty of measurement of phase place by the 4th uncertainty of measurement that goes on foot the S parameter that obtains;

Then, according to the uncertainty of measurement of mould and the uncertainty of measurement of phase place, calculate the standard handovers gain G with reference to formula (9) _T(S) uncertainty of measurement;

At last, calculate robot scaling equipment because the uncertainty of measurement that measuring repeatability is introduced, with itself and G _T(S) the synthetic standard handovers gain G that obtains of uncertainty of measurement _T(S) calibration uncertainty.

Images