Probing for oscilloscope

Probing for oscilloscope

Agenda - Notion de sonde en oscilloscopie - Structure des différentes sondes - Passives - Actives - Logiques - Différentielles - Comment choisir la bonne sonde - Nouvelle technologie pour sonde différentielle

Sonde, notions Voltage Logic Current Optical Passive Active Passive Active Passive Active Z0 High Z Differential High Voltage Single- Differential Ended AC DC AC

Sonde, Intégrité et fidélité Original signal Signal at probe tip is changed Signal at the oscilloscope Signal display with DSP correction

Sonde Idéale R in = C in = 0 CABLE BANDWIDTH = The ideal probe would have no effect on the signal being measured, zero loading. Infinite Bandwidth Zero Input Capacitance Infinite Input Resistance Attenuation of 1 Zero Delay Zero Phase Shift Mechanically well suited to application Infinite Dynamic Range

Sonde Réelle C C V CC R C Without probe Gain = - R C R E f 0 = 1 2 R C C C V IN PROBE R P C P With probe Gain = - (R C R P ) f 0 = R E 1 2 (R C R P )(C C +C P ) R E NOTE: V CC is an AC Ground

Sonde Réelle Impédance R in acts like a voltage divider Higher input resistance less loading Lower source resistance less loading DUT Probe R source Probe Tip R In E source L Ground Lead C In Decreased Signal Amplitude V meas = V source R in R in + R source 100% 90% 100% 90% 10% 0% 10% 0% Source Signal Effects of Input Resistance

Sonde Réelle Charge

Sonde Réelle Inductance

Sonde Réelle Inductance Masse

Amplitude Sonde Réelle BW/Rise Time 3% 10% Frequency 20% 30% The signal delivered to the oscilloscope Insufficient BW degrades the front edge Follow the 1/5 th Rule T system < T r,signal T r,measured = (T r,signal ) 2 + (T r, system ) 2 5 Actual Measured

1X Probe Model (Length of Cable) DUT PROBE SCOPE R source V source Probe Tip L Ground Lead PROBE CABLE 8-10 pf/ft * 1.5 ns/ft 6 feet 1 MΩ 20 pf Advantages: 1X (No Attenuation) Inexpensive Disadvantages: Very High Reflections Very High Input C Very Low Bandwidth * Typical 50 Ω cable has about 30 pf/ft of capacitance

Typical High Z 10X Passive Probe Model DUT C1 8-12 pf PROBE SCOPE R source V source Advantages: Probe Tip 9 MΩ R1 High Input R Wide Dynamic Range Inexpensive Mechanically Rugged Low Input C vs 1X Probe L Ground Lead PROBE CABLE 8-10 pf/ft 1.5 ns/ft 6 feet 500 Ω C3 7-50 pf Disadvantages: Input C Too High R3 Not Compatible with 50 Ω Systems Must be Compensated 1 MΩ R2 C2 20 pf

50 Ω Divider Probe (Z0) Model (10X) DUT 450 Ω PROBE SCOPE R source V source 0.5 pf Probe Tip L Ground Lead PROBE CABLE 50 Ω 50 Ω 6 feet Advantages: Low Input C Wide Bandwidth to 9 GHz Compatible with 50 Ω Systems and 1 MΩ with Termination Resistor No Compensation Necessary Disadvantages: Low Input R Must be Terminated into 50 Ω

Active Probe Model DUT PROBE 6 feet SCOPE R source V source Probe Tip L Ground Lead BUFFER AMP PROBE CABLE 50 Ω R t 50 Ω 1 M Ω / 50 Ω Advantages: Low Input Capacitance Wide Bandwidth High Input R Compatible with 50Ω Systems or 1 MΩ with Termination Resistor R t No Compensation Necessary Disadvantages: Higher Cost Limited Dynamic Range Mechanically Less Rugged Requires Power

Active Differential Probes PROBE SCOPE + _ V OUT CH1 Scope Amplifier Typical CMRR 10,000 : 1 @ DC 2000 : 1 @ 20 MHz Advantages: Lower Input Capacitance Higher CMRR vs Frequency Than Passive Differential Pair Compatible With 50 and 1 M Single-ended Systems Disadvantages: Higher Cost Limited Dynamic Range Mechanically Less Rugged and Larger Size Requires Power

TPP1000 TPP0500 TPP0502 Passive Probes

Sondes Trimode 30GHz

TriMode Probing TriMode, with a single probe-dut connection, allows: Traditional differential measurements: V+ to V- Independent single ended measurements on either input V+ with respect to ground V- with respect to ground Direct common mode measurements: (V+) + (V-)/2 with respect to ground Serial Data standards such as PCI Express, Serial ATA, etc require both differential and maximum permissible common mode voltage limit measurements. Requires two separate probes --- Until Now! 19

Sonde, le bon choix

Probing Impedance loading

Input Impedance 100000 P73xx Input C ~ 0.21pF 10000 TriMode P75xx Avec charge modification forme du signal Input Impedance (Ohms) 1000 1169A P7516 P6150 100 Input C ~ 0.05pF` Pas de charge Pas de modification forme du signal Impédance Flat 10 1.00E+04 10K 1.00E+05 100K 1.00E+06 1M 1.00E+07 10M 1.00E+08 100M 1.00E+09 1G 1.00E+10 10G 1.00E+11 100G Frequency (Hz) 9/19/2016 TEKTRONIX CONFIDENTIAL 22

ISOVu, nouvelle technologie

What Limits Power Measurements? THE PROBE S INABILITY TO DELIVER AN ACCURATE SIGNAL No probe today has the combination of bandwidth, common mode voltage and common mode rejection Today s probes derate over bandwidth Voltage, loading and CMRR Long leads make the probe susceptible to noise and radiated emissions The waveform shape changes based on the lead dress and the position of the probe head Large hook clips are inadequate for dense environments and lack performance Susceptible to ground loops 19 SEPTEMBER 2016 24

Voltage Derates Over Bandwidth YOU DON T GET THE FULL DYNAMIC RANGE AT FULL BANDWIDTH The specified dynamic range is only true for DC and low frequencies This 100 MHz 1 kv RMS probe starts to roll off at 2 MHz 2 MHz 19 SEPTEMBER 2016 25

Understanding the Measurement Issue THE EFFECT OF COMMON MODE VOLTAGE A V A-B V Out V Diff B V CM 40 V A V Diff ~0V 1V V Diff : 0 V 1 V V CM : 40 V B 40 V 19 SEPTEMBER 2016 26

CMRR Derates Over Bandwidth POWER MEASUREMENTS REQUIRE A PROBE WITH HIGH CMRR 19 SEPTEMBER 2016 27

Designing Blind THE EFFECT OF COMMON MODE VOLTAGE Gate Driver Vdiff mvs to ± 50V Conventional differential probing systems don t have the required combination of bandwidth and common mode rejection Vcm < 2000V Engineers are forced to rely on simulation, workarounds, or inferences (forced to measure to ground). 19 SEPTEMBER 2016 28

What is IsoVu Technology? ISOLATED MEASUREMENT SYSTEM IsoVu technology is a radically new measurement system platform Enables high-resolution measurements on differential signals Utilizes an electro-optic sensor to convert the input signal to optical modulation Galvanically isolates the device-under-test Incorporates: 4 separate lasers Optical sensor Power over fiber 5 optical fibers Sophisticated feedback and control techniques 19 SEPTEMBER 2016 29

IsoVu Doesn t Derate Over Bandwidth YOU GET THE FULL DYNAMIC RANGE AT FULL BANDWIDTH IsoVu is the only product in the world with a flat derating 19 SEPTEMBER 2016 30

IsoVu Offers 1 Million to 1 CMRR ISOVU IS 100,000 TIMES BETTER THAN STANDARD SOLUTION 1 Million to 1 at 100 MHz More than 10,000 to 1 at 1 GHz 19 SEPTEMBER 2016 31

High-Side Gate Measurements OPTIMIZATION AND TUNING MAY BE IMPOSSIBLE Three characteristic regions during turn on: 1. C GS : Charge Time 2. Miller Plateau: Time required to charge the gate-drain Miller capacitance (C GD ) 3. Conduction: Gate charges to its final value 19 SEPTEMBER 2016 32

Making Signal Details Visible 1 GHZ BANDWIDTH, 2000 V CM AND 1,000,000:1 CMRR LeCroy DA1855A IsoVu Gate Driver Vdiff mvs to ± 50V IsoVu rejects common mode noise, so you can see small differential signals in the presence of common mode voltages Vcm < 2000V Engineers can confirm simulations and evaluate signal characteristics, such as this ringing. 19 SEPTEMBER 2016 33

Technologie Optique Mach-Zehnder 19 SEPTEMBER 2016 34

IsoVu Overview ISOLATED HIGH VOLTAGE MEASUREMENT SYSTEM Scope 50Ω Term Fiber optic isolation Power over Fiber (no batteries required) 3 meter and 10 meter options TekVPI Interface Controller Optical Fiber Sensor Head Probe Tip 19 SEPTEMBER 2016 35

Merci