Passive Component Analysis. OMICRON Lab Webinar Nov. 2015

Transcription

1 Passive Component Analysis OMICRON Lab Webinar Nov. 2015

2 Webinar Hints Activate the chat function Please mute your microphones to avoid echoes Feel free to post questions anytime using the chat function Please post your questions to Bernhard Baumgartner We will record the presentation such that you can view it again later Please mute yourself by clicking on this icon! Send questions via chat to Bernhard Baumgartner Page 2

3 Agenda Why do we analyze passive components How to measure component impedance A detailed look at a capacitor Inductor and transformer Filter simulation vs. real world Summary Page 3

4 Passive Components Essential parts in analog circuits Inductor and capacitor used e.g. to store energy or to create filter circuits Inductor: v t = L di t dt X L = ωl V I = Z L = jωl Capacitor: i t = C dv t dt X C = 1 ωc V I = Z C = 1 jωc Page 4

5 Theory and Reality Theoretically inductor and capacitor are purely reactive elements No resistive behavior and therefore lossless In reality parasitics can strongly influence the real behavior especially at higher frequencies Examples: Inductor: Wire has resistance Windings form electric field Core is not lossless Capacitor: Plates are resistive Rolling of foils creates inductance Insulator not lossless Page 5

6 TR2/H Equivalent Circuits Are used to model the real behavior of the components Different complexity of models 1 st order models are valid for one frequency Single Frequency Mode in BAS calculates R, L and C Frequency Sweep Mode calculates R, L and C over frequency 10 10n 8 8n TR1/Ohm n 4n 2n 0 0 5M 10M 15M 20M 25M 30M 35M 40M f/hz TR1: Rs(Impedance) TR2: Ls(Impedance) Page 6

7 Equivalent Circuits Higher complexity models are valid for a frequency range 2 nd Order equivalent circuits for inductor and capacitor 3 rd Order models (e.g. quartz crystal or piezo element) see Application Note: Equivalent Circuit Analysis of Quartz Crystals Parameter identification requires manual work or e.g. curve-fitting procedure Page 7

8 Bode 100 Impedance Measurement Methods Direct Measurements One-Port Reflection Impedance Adapter (3-port technique) External bridge (e.g. high impedance bridge) Indirect Measurements (via Gain) Shunt-Thru (2-port technique) Series-Thru (2-port technique) Voltage-Current Gain (3-port technique) Page 8

9 Direct Measurement Methods One-Port Impedance Adapter Recommended for 0.5 Ω - 10 kω Recommended for 20 mω kω External Bridge / Coupler Range depends on bridge Page 9

10 Indirect Measurement Methods Shunt-Thru S 21 Z DUT = 25Ω 1 S 21 Series-Thru Z DUT = 100 Ω 1 S 21 S 21 Recommended for 1 mω - 10 Ω Recommended for 1 kω < 10 MΩ Voltage Current Gain Gain = V CH2 V CH1 = V I = Z DUT Page 10

11 Impedance Range Overview Page 11

12 Why is it important to measure capacitors? A capacitor is NEVER just a capacitor Capacitor ESR influences the phase margin of power supplies Capacitor ESR influences the output ripple at the switching frequency of a SMPS ESR can change over Frequency Capacitors are inductors above their resonance frequency Page 12

13 What does the data sheet tell us? 220 µf aluminum capacitor C = 220µF (± 20%) ESR = tan δ ωc = π 120Hz 220µF = Hz Page 13

14 This is what the measurement tells us TR1/Ohm Impedance f/hz TR1: Mag(Impedance) TR1/Ohm f/hz TR1/Ohm Cursor 1 120, ,077m TR1: Rs(Impedance) f/hz ESR TR2/ 0-50 Phase f/hz TR2: Phase(Impedance) Capacity Page 14

15 Calibration Open Short Load Page 15

16 User Calibration / Probe Calibration User Calibration (User Range Calibration) Calibrates at exactly the frequencies that are currently measured + No interpolation, suitable for narrowband probes Probe Calibration (Full Range Calibration) calibrates at pre-defined frequencies and interpolates in-between + Calibration does not get lost when frequency range is changed Page 16

17 Detailed Example available TR2/F 5m 4m 3m 2m 1m 0-1m -2m -3m -4m -5m f/hz TR2: Cs(Impedance) 10 0 TR1/Ohm TR1: Rs(Impedance) f/hz see Application Note: Capacitor ESR Measurement with Bode 100 and B-WIC Page 17

18 Fitting Model to Measured Impedance Various methods available We use curve-fitting A Preview tool is available on request Page 18

19 TR2/ Simulation vs. Measurement TR1/Ohm TR1: Mag(Impedance) f/hz TR2: Phase(Impedance) Page 19

20 Voltage sensitivity of capacitors see Application Note: DC Biased Impedance Measurements Page 20

21 Why should we measure inductors? An inductor is NEVER just an inductor AC resistance <> DC resistance skin effects Eddie Currents Inductors have resonance frequencies Inductors with magnetic cores can have core losses Page 22

22 What does the data sheet tell us? 33 µh shielded power inductor H = 33µH (± 1 khz R DC =0,049 Ω (typ.) R DC =0,057 Ω (max.) f res = 11 MHz Page 23

23 This is what the measurement tells us TR2/ TR1/Ohm f/hz TR1/Ohm 10 4 Cursor 1 13,362M 40,673k TR1: Mag(Impedance) 200 f/hz TR2/ 150 Cursor 1 13,362M -5, TR2: Phase(Impedance) f/hz f/hz Impedance Phase TR1/Ohm TR2/H f/hz TR1/Ohm 10 4 Cursor 1 292,486 40,049m Cursor 2 101,789k 759,543m 10 3 C2-C1 101,496k 719,494m u f/hz TR2/H Cursor 1 292,486 33,019µ Cursor 2 101,789k 30,551µ 200u C2-C1 101,496k -2,468µ 100u 0-100u TR1: Rs(Impedance) u TR2: Ls(Impedance) f/hz f/hz Inductance ESR Page 24

24 Flyback Transformer Leakage Inductance Not all flux generated by the primary winding is coupled to the secondary winding some flux leaks some contributes to core losses Represented by a series inductance in the circuit Leakage inductance creates a voltage spike when turning off current through primary side (flyback converter) Page 25

25 Measuring Leakage Inductance Leakage inductance is measured by shorting all other windings except the primary winding Secondary open Secondary shorted Leakage inductance TR1/Ohm TR2/H 12u 10u 8u 6u 4u 2u TR1: Mag(Impedance) f/hz Memory 1 : Mag(Impedance) TR2: Ls(Impedance) Leakage inductance is not constant over frequency f/hz Page 26

26 LC Filter Bode Diagram Simulation in LTSpice: Passband Resonance (Double-pole) Stopband -40dB/Decade -180 Phase Page 27

27 LC Filter Test board Measuring the voltage transfer function H jω = V out V in Page 28

28 Measurement vs. Simulation Measurement Simulation Stopband is different Phase does not reach -180 Second resonance at 30 MHz parasitic effects Page 29

29 TR2/ LC Filter Including Parasitics TR1/dB TR1: Mag(Gain) f/hz TR2: Phase(Gain) much better fit between simulation and measurement Could be further improved by better component models Page 30

30 Reducing Output Ripple 2 x 10µF ceramics adds 20dB attenuation at 300 khz TR1/dB f/hz TR1/dB Cursor 1 290,070k -62,575 Imroved stop band performance at 300 khz (e.g. switching frequency) TR1: Mag(Gain) f/hz Memory 1 : Mag(Gain) Page 31

31 Summary Component parasitics are important to understand real life circuit behavior Models considering parasitics allow better simulation Measuring components can tell us more than the data sheet says Page 32

32 Feel free to ask questions via the chat function... If time runs out, please send us an and we will follow up. You can contact us at: Thank you for your attention!