Jan Verspecht bvba Gertrudeveld 15 1840 Steenhuffel Belgium - PowerPoint PPT Presentation

Jan Verspecht bvba Gertrudeveld 15 1840 Steenhuffel Belgium email: contact@janverspecht.com web: http://www.janverspecht.com Calibration of a Measurement System for High Frequency Nonlinear Devices Jan Verspecht Slides of the Doctoral Dissertation - Vrije Universiteit Brussel, November 1995

1 Calibration of a Measurement System for High Frequency Nonlinear Devices Jan Verspecht

2 Overview • Introduction • Vectorial “Nonlinear Network” Analyzer Hardware • Accuracy of Broadband Sampling Oscilloscopes • The “Nose-to-Nose” Calibration Procedure • Absolute Calibration of a VNNA • Consistency Check: Model versus Measurements • Conclusions

3 • Introduction • Vectorial “Nonlinear Network” Analyzer Hardware • Accuracy of Broadband Sampling Oscilloscopes • The “Nose-to-Nose” Calibration Procedure • Absolute Calibration of a VNNA • Consistency Check: Model versus Measurements • Conclusions

4 The High-Tech World

5 Engineering Tools Prototyping Producing Measuring Idea Computing

6 High-Speed Electronic Measurements Filters Dynamic Nonlinear TDR-Oscilloscopes Small signal amplifiers Network Analyzers Interconnects Dynamic Linear “Nonlinear Network” Analyzer Mixers Frequency Multipliers Spectrum Analyzers Power Amplifiers Oscilloscopes Digital Static Nonlinear

7 VNNA Measurements Vectorial “Nonlinear Network” Analyzer Amplitude & Phase a 2 a 1 input input DUT output output b 1 b 2 Nonlinear

8 Instrument Calibration VNNA Scientific and Industrial World Ω kg s H W m V A K Hz F J

9 • Introduction • Vectorial “Nonlinear Network” Analyzer Hardware • Accuracy of Broadband Sampling Oscilloscopes • The “Nose-to-Nose” Calibration Procedure • Absolute Calibration of a VNNA • Consistency Check: Model versus Measurements • Conclusions

10 VNNA Hardware Preexistent Prototypes Broadband Oscilloscopes Couplers Linear Network Analyzer Test Sets Synthesizers RF Switches Amplitude Cal RF Power Meter Measurements Phase Cal ? Amplitude & Phase Ideal RF signal samplers Incident & Reflected “Golden diode” Port 1 & Port 2

11 HP-NMDG Prototype precision analog-to-digital convertor Spec Sheet Bandw.: 18GHz 4 channel broadband downconvertor Dyn. Range: 60dB DUT BIAS1 BIAS2 (on wafer)

12 Traceability Of Calibration Amplitude Cal Standards Lab RF Power Meter Phase Cal “Nose-to-Nose” Calibration Broadband Sampling Oscilloscope Reference Waveform Generator

13 • Introduction • Vectorial “Nonlinear Network” Analyzer Hardware • Accuracy of Broadband Sampling Oscilloscopes • The “Nose-to-Nose” Calibration Procedure • Absolute Calibration of a VNNA • Consistency Check: Model versus Measurements • Conclusions

14 Sampling Oscilloscope Basics Trigger Level Delay Setting DAC DELAY TRIGGER ∫ CH1 Screen ∫ CH2 ∫ CH3 ∫ CH4

15 Timebase Errors HLog estimator Drift Log Spectral Averaging Noise rms Jitter = F(signal derivative) Distortion Phase Demodulation

16 Vertical Errors Gain DC measurements Offset Limit Signal Distortion Amplitude Dynamical “Nose-to-Nose” Charact.

17 • Introduction • Vectorial “Nonlinear Network” Analyzer Hardware • Accuracy of Broadband Sampling Oscilloscopes • The “Nose-to-Nose” Calibration Procedure • Absolute Calibration of a VNNA • Consistency Check: Model versus Measurements • Conclusions

18 Impulse Response = Kick-Out Dirac Delta Impulse Response + Precharged - Kick-Out

19 “Nose-to-Nose” Measurement + - Measured Kick-Out

20 Systematic Measurement Error Fourier transform of sampling aperture waveform measured freq. response function ) e j ϕ P ω ( ( ) ) M ω ( ) H ω ( = MEASUREMENT ERROR physical freq. response function

21 Upperbound For Phase Error Sampling Aperture : Max. Error ( degrees) 12 1 Local Max 10 8 Positive 6 4 2 0 0 10 20 30 40 50 10ps Frequency (GHz) Finite In Time

22 Comparison With Power Measurements 0 High BW High BW -3 -3 Response (dB) Response (dB) Low BW ow BW -6 -6 -9 -9 -12 -12 0 10 10 30 30 20 20 40 40 50 50 60 60 Frequency (GHz) requency (GHz)

23 Repeatability and Noise Amplitude Phase 1.5 0.2 Amplitude Diff. (dB) 1 Phase Diff. (deg) 0.1 0.5 0 0 -0.5 -0.1 -1 -0.2 -1.5 0 10 20 30 40 50 0 10 20 30 40 50 Frequency (GHz) Frequency (GHz)

24 Sampler Linearity amplitude differences (dB) frequency (GHz)

25 Frequency Response Function Amplitude Phase 0 15 -1 Amplitude (dB) 12.5 Phase (deg) -2 10 -3 7.5 5 -4 2.5 -5 0 -6 5 5 10 15 20 25 30 10 15 20 25 30 Frequency (GHz) Frequency (GHz)

26 “Nose-To-Nose” Errors (Frequency = 20GHz) Amplitude Phase Time Distortion 8mdB 0.02deg Timing Jitter 150mdB / Time Drift / / Repeatab., Noise 20mdB 0.15deg Nonlinearity < 20mdB < 0.15deg Unknown Phase / 0.72deg

27 • Introduction • Vectorial “Nonlinear Network” Analyzer Hardware • Accuracy of Broadband Sampling Oscilloscopes • The “Nose-to-Nose” Calibration Procedure • Absolute Calibration of a VNNA • Consistency Check: Model versus Measurements • Conclusions

28 Error Model a m1 b m1 a m2 b m2 a d1 a d2 a g1 a g2 LINEAR LINEAR DUT b g1 b g2 b d1 b d2 Definition of Variables

29 Calibration powermeter i i i 1 β 1 0 0 a m 1 a d 1 i δ 1 i i i γ 1 0 0 b m 1 b d 1 K i e j ϕ K i ( ) = i β 2 i i i 0 α 2 0 a m 2 a d 2 i δ 2 i i i 0 γ 2 0 b m 2 b d 2 reference generator classical calibration (LOS, LRM)

30 Absolute Calibration Amplitude power sensor PORT 1 Phase reference gen. PORT 1

31 On Wafer: Reciprocity power sensor Data Acquisition Line reference gen. probe tip RECIPROCITY APC-3.5

32 The Reference Generator trigger signal 1.5m SMA cable differentiator APC-3.5 conn. (m) 1GHz amplifier SRD module 20dB att. (0.5W) 240 180 Amplitude (mV) 120 60 0 -60 -120 -180 -240 200 400 600 800 1000 Time (ps)

33 Absolute Calibration Factor Amplitude Phase 21 -20 Amplitude (dB) 20 -25 Phase (deg) 19 -30 18 -35 17 -40 16 -45 3 6 9 12 15 18 3 6 9 12 15 18 Frequency (GHz) Frequency (GHz)

34 • Introduction • Vectorial “Nonlinear Network” Analyzer Hardware • Accuracy of Broadband Sampling Oscilloscopes • The “Nose-to-Nose” Calibration Procedure • Absolute Calibration of a VNNA • Consistency Check: Model versus Measurements • Conclusions

35 Model vs. Measurements DUT s-parameters as a function of DC bias Large Signal Model (Root or Jansen et al. ) ? = Simulation VNNA Data

36 Early Results MESFET transistor (INTEC-UG, IMEC): HDA, fundamental 3GHz : Root-model : VNNA measurements

37 More Advanced Results HEMT transistor (ESAT-KUL, IMEC): HDA, fundamental 3GHz 0 Harmonic output power (dBm) -10 -145 Phase (deg) -150 -20 -155 -30 -160 -165 -40 -25 -20 -15 -10 Input power (dBm) -50 : Jansen et al. model -60 : VNNA measurements -5 -25 -20 -15 -10 Input power (dBm)

38 Time Domain 0.1 0.05 Amplitude (V) 0 -0.05 -0.1 -0.15 0 100 200 300 400 500 600 Time (ps) : VNNA measurement : Jansen et al. model : incident voltage wave

39 • Introduction • Vectorial “Nonlinear Network” Analyzer Hardware • Accuracy of Broadband Sampling Oscilloscopes • The “Nose-to-Nose” Calibration Procedure • Absolute Calibration of a VNNA • Consistency Check: Model versus Measurements • Conclusions

40 Conclusions • The Instrumentation And Calibration Procedure Developed Allows the Accurate Measurement of Phase and Amplitude of the Spectral Components of Incident and Scattered Voltage Waves at the Signal Ports of a Nonlinear Microwave Device. • The Relative Calibration And Amplitude Calibration Are Traceable To National Standards, The Phase Calibration Is Traceable To The “Nose-To-Nose” Calibration Procedure.

41 Future Research • Phase Calibration Traceable To National Standards • More Theoretical Work Concerning “Nose-To-Nose” • Implementing The “Nose-To-Nose” For Photo-Conductive Samplers • Putting Error Flags On VNNA Measurements • Towards Commercial Use Of VNNA