METHOD FOR CALIBRATING A TEST RIG

摘要

A method for calibrating a test apparatus, having a first and a second directional coupler, for gauging a two-port test object that has a first port and a second port in a calibration plane, wherein for the purpose of calibrating the test apparatus a vectorial network analyzer having a 1st-6th test port is connected to the first and second ports in the calibration plane such that the first and second test ports are connected to respective port in the calibration plane, the third and fourth test ports are connected to the first directional coupler and the fifth and sixth test ports are connected to the second directional coupler via a respective waveguide for electromagnetic waves. For different calibration standards, scatter parameters are determined for each desired frequency point. For the different calibration standards, corrections to the scatter matrix are made in order to obtain a corrected scatter matrix. The scatter parameters of the corrected scatter matrix are used to determine terms for error matrices.

主权项

1. A method for calibrating a test apparatus, having a first and a second directional coupler, for testing a two-port test object (DUT—Device Under Test), which has a first port and a second port in a calibration plane;
wherein for the purpose of calibrating the test apparatus a vectorial network analyzer (VNA) having a first, a second, a third, a fourth, a fifth and a sixth test port is connected to the first and second ports in the calibration plane such that the first test port is connected to the first port in the calibration plane, the second test port is connected to the second port in the calibration plane, the third and fourth test ports are connected to the first directional coupler and the fifth and sixth test ports are connected to the second directional coupler via a respective waveguide for electromagnetic waves; wherein at the first test port an electromagnetic wave a1 runs out in the direction of the first port in the calibration plane and an electromagnetic wave b1 runs in from the direction of the first port in the calibration plane; wherein at the second test port an electromagnetic wave a2 runs out in the direction of the second port in the calibration plane and an electromagnetic wave b2 runs in from the direction of the second port in the calibration plane; wherein at the first port in the calibration plane an electromagnetic wave aDUT,1 runs in from the direction of the first test port and an electromagnetic wave bDUT,1 runs out in the direction of the first test port; wherein at the second port in the calibration plane an electromagnetic wave aDUT,2 runs in from the direction of the second test port and an electromagnetic wave bDUT,2 runs out in the direction of the second test port; wherein between the first test port and the first port in the calibration plane a component of the wave a1 is coupled out by the first directional coupler as aMess,1 and fed to the third test port of the VNA; wherein between the first test port and the first port in the calibration plane a component of the wave b1 is coupled out by the first directional coupler as bMess,1 and fed to the fourth test port of the VNA; wherein between the second test port and the second port in the calibration plane a component of the wave a2 is coupled out by the second directional coupler as aMess,2 and fed to the fifth test port of the VNA; wherein between the second test port and the second port in the calibration plane a component of the wave b2 is coupled out by the second directional coupler as bMess,2 and fed to the sixth test port of the VNA; wherein in order to calibrate the test apparatus, instead of the DUT at least three different calibration standards are arranged in the calibration plane; wherein for each calibration standard K and for each desired frequency point of a frequency f of a1 or a2 scatter parameters Sxy,K,f, where x=1, 2, 3, 4, 5 or 6 and y=1 or 2, are determined between the y-th and the x-th test port of the VNA for the calibration standard K and the frequency f is determined from the known values a1,K,f and a2,K,f as well as from the measured values b1,K,f, b2,K,f, aMess,1,K,f, bMess,1,K,f, aMess,2,K,f, bMess,2,K,f, whereby[b1,K,fb2,K,f]=[S11,K,fS12,K,fS21,K,fS22,K,f][a1,K,fa2,K,f][aMess,1,K,fbMess,1,K,f]=[S31,K,fS32,K,fS41,K,fS42,K,f][a1,K,fa2,K,f][aMess,2,K,fbMess,2,K,f]=[S51,K,fS52,K,fS61,K,fS62,K,f][a1,K,fa2,K,f] wherein a scatter matrix Sunkorr,K,fSunkorr,K,f=[S11,unkorr,K,fS12,unkorr,K,fS21,unkorr,K,fS22,unkorr,K,f], describing the transmission via the directional coupler is calculated from the measured scatter parameters Sxy,K,f of the calibration standards, where x=3, 4, 5, 6 and y=1, 2, with scatter parameters S11,unkorr,K,f, S12,unkorr,K,f, S21,unkorr,K,f and S22,unkorr,K,f, according toS11,unkorr,K,f=bMess,1,K,faMess,1,K,f=S42,K,fS32,K,f=σ11S21,unkorr,K,f=bMess,2,K,faMess,2,K,f=S61,K,fS31,K,f=σ21S12,unkorr,K,f=bMess,1,K,faMess,2,K,f=S42,K,fS32,K,f=σ12S22,unkorr,K,f=bMess,2,K,faMess,2,K,f=S62,K,fS52,K,f=σ22 whereby a scatter matrix SI,K,f describing the transmission between the first test port of the VNA and the first port in the calibration plane on the one hand, as well as between the second test port of the VNA and the second port in the calibration plane on the other hand, is determined with the measured scatter parameters Sxy,K,f of the calibration standards, where x=1, 2 and y=1, 2, asSI,K,f=[S11,K,fS12,K,fS21,K,fS22,K,f] whereby, with the measured scatter parameters Sxy,K,f of the scatter matrix SI,K,f, where x=1, 2 and y=1, 2, teens i00, i01·i10 and i11 of an error matrix IA withIA=[i00i01i10i11] are determined for a signal transmission between the first test port on the one hand and the first port in the calibration plane on the other hand depending on a frequency f of the waves a1 or a2 by means of a predetermined calibration algorithm, where IA is a scatter matrix according to[b1aDUT,1]=IA[a1bDUT,1] whereby, with the measured scatter parameters Sxy,K,f of the scatter matrix SI,K,f, where x=1, 2 and y=1, 2, terms i22, i23·i32 and i33 of an error matrix IB withIB=[i22i23i32i33] are determined for a signal transmission between the second test port on the one hand and the second port in the calibration plane on the other hand depending on a frequency f of the waves a1 or a2 by means of a predetermined calibration algorithm, whereby IB is a scatter matrix according to[aDUT,2b2]=IB[bDUT,2a2] such that for each frequency step with the frequency f of a1 or a2 and for each calibration standard K a correction of the scatter matrix Sunkorr,K,f is carried out to produce a corrected scatter matrix Sc,K,f according to the formulaSc,K,f=1D[σ11-σ12σ21ΓF,K,fσ12-σ11σ12ΓR,K,fσ21-σ22σ21ΓF,K,fσ22-σ12σ21ΓR,K,f] where D=1−σ12σ21ΓF,K,fΓR,K,f, andΓF,K,f=S51.K,fS61.K,fdescribes the ratio of propagated to reflected wave measured at the output of the second directional coupler on feeding through the first test port of the VNA andΓR,K,f=S32.K,fS42.K,fdescribes the ratio of propagated to reflected wave measured at the output of the first directional coupler on feeding through the second test port of the VNA;
whereby, with the scatter parameters of the scatter matrix Sc,K,f, terms e00, e01·e10 and e11 of an error matrixEA=[e00e01e10e11] are determined for a signal transmission between the third and fourth test port on the one hand and the first port in the calibration plane on the other hand depending on a frequency f of the waves a1 or a2 by means of a predetermined calibration algorithm, where EA is a scatter matrix according to[bMess,1aDUT,1]=EA[aMess,1bDUT,1] whereby, with the scatter parameters of the scatter matrix Sc,K,f, terms e22, e23·e32 and e33 of an error matrixEB=[e22e23e32e33] are determined for a signal transmission between the fifth and sixth test port on the one hand and the second port in the calibration plane on the other hand depending on a frequency f of the waves a1 or a2 by a predetermined calibration algorithm, whereby EB is a scatter matrix according to[aDUT,2bMess,2]=EB[bDUT,2aMess,2] whereby the isolated terms i01 and i10 from the product i01·i10 as well as the isolated terms i23 and i32 from the product i23·i32 are determined according to the formulas
i01=i10=±√{square root over (i01i10)}andi23=i32=±√{square root over (i23i32)} whereby the sign is in each case determined starting out from a frequency point with known phase through continuous extrapolation, whereby a phase difference from one frequency point to a next frequency point is reduced by 180° if this phase difference exceeds a predetermined threshold value; whereby the isolated term e10, is calculated from the product e10·e01 according to the formulae10=i10·a1aMess,1·1-e11S11,DUT,K*1-i11S11,DUT,K* and the isolated term e01 is determined from this, wherebyS11,DUT,K*=(S11,K*,f-i00)(i10·i01+i11·(S11,K*,f-i00)) and K* designates a calibration standard without transmission; whereby the isolated term e23 from the product e32·e23 is calculated according to the formulae23=i23·a2aMess,2·1-e22S22,DUT,K*1-i22S22,DUT,K* and the isolated term e23 is determined from this, wherebyS22,DUT,K*=(S22,K*,f-i22,f)(i32·i23+i33·(S22,K*,f-i22,f)) and K* designates a calibration standard without transmission.