HIP @ ICE

using high-impedance probesfor in-circuit voltage

time-domain measurements

2/2012

© NMDG

© NMDG 2012 2

Summary

Goal: accurate in-circuit measurement of time-domain voltage using one or more high-impedance probes (HIP) within a large-signal network analysis context @ locations normally unreachable using a vector network analyzer

Supported: Multiple probes Probing on PCB Probing on wafer

What are the voltage waveforms in-between the components?

How to measure these accurately? Apply proper calibration techniquesHow to minimize their disturbance? Use high-impedance probes

© NMDG 2012 3

Outline

Calibrated measurement using “home-made” HIP and ZVxPlus (*)

Measurement setup Measurement results Impact of high-impedance probing Calibration

Calibrated measurement using commercial Kelvin probe and real-time scope Measurement setup Measurement results Impact of high-impedance probing Calibration

(*) R&S ZVA24 VNA with nonlinear extensions to allow time-domain measurements

© NMDG 2012 4

Measurement using calibrated HIP in practice (I)Sanity check measuring the output voltage of an E-pHEMT FET

-5 dBm input @ 250 MHz, Vgs = 0.4 V, Vds = 4 V

2nd calibrated HIP probe allows to measure the input voltage

2nd

probeprobe

source 50Ωtermination

P1 P2P4synchronizer P3

probes

© NMDG 2012 5

Measurement using calibrated HIP in practice (II)

v2(t)excellent agreement between output voltage (red)measured using the TRL-calibrated coupler-basedtwo-port network analyzer and the voltage (green)measured by the calibrated high-impedance probe

v2(f)corresponding excellent agreement

of amplitude and phase in thefrequency domain

upper: coupler-based amp & phaselower: calibrated HIP amp & phase

© NMDG 2012 6

Measurement using calibrated HIP in practice (III)“a global ICE perspective”

© NMDG 2012 7

Impact of high-impedance probing(*)

The impedance of the probe is measured during calibration↓

The impact of the high-impedance probe can be taken into account

Estimated impedance of “home-made” (2k4) HIP

0.5 1 1.5 2

1400

1600

1800

2000

2200

2400

2600 0.5 1 1.5 2

60

50

40

30

20

10

f (GHz)

f (GHz)| ZHIP |

ZHIP (deg)

(*) in fact the combination of the probe, cable and receiver input

© NMDG 2012 8

Including impact of high-impedance probe during simulation

Probe

including the load effect of the probe in the simulation to compare measurements with simulations

v1i1

i2v2v probe

Figure and schematic courtesy of AWR

© NMDG 2012 9

Calibration process

Step 1: perform a “standard” one-port or two-port VNA calibration, extended with power and phase calibration to measure time-domain information

Step 2: use the calibrated system resulting from step 1 to calibrate the high-impedance probe

Step 3: use the calibrated high-impedance probe to measure the time-domain voltage in-circuit

© NMDG 2012 10

Calibration of high-impedance probe (step 2)

Different options are possible, such as :

Starting from an absolute TRL-based two-port calibration : Connect the HIP to the open of the TRL calibration kit (fig. 1) Connect the HIP to the thru of the TRL calibration kit (fig. 2)

Starting from an absolute on-wafer one-port calibration : Connect the HIP to the open of the on-wafer calibration kit

Starting from an absolute coaxial one-port (or two-port) calibration : Connect the HIP to the open on a microstrip or coplanar waveguide PCB

provided by NMDG De-embedding information provided with PCB allows to move the absolute

calibration from the coaxial plane to the open on the PCB

fig. 1 fig. 2

© NMDG 2012 11

High-impedance probe calibration in practice (step 2)

Connecting the “home-made” (2k4) HIPto the thru of the TRL calibration kit

© NMDG 2012 12

HIP and real-time scope : ESD (TLP) @ OnSemi

Comparing reconstructed voltage waveforms at the DUT, measured on a 7.3VZener diode, subjected to a 200V TLP pulse. Blue : reconstructed using data from the directional couplers. Red : corrected data from Kelvin-probe.

Photos and figure courtesy of Renaud Gillon (OnSemi, Oudenaarde, Belgium)

Using a calibrated Kelvin probe as part ofa transmission line pulse (TLP) system tore-construct dynamic voltage waveforms on low-impedance devices subjected tohigh currents with a very significant improvement in the signal-to-noise ratio

© NMDG 2012 13

TLP measurement setup

charge line

Tek scopeDP070404

cable1 cable2

vD

7.3 V Zener diode

x3m cable3

Kelvinprobe

http://www.ggb.com/10.html

200 V, 25 ns pulse0.6 ns rise time

WerlatoneC6493 30 dB

coupler

Kelvin probe

© NMDG 2012 14

Combined impact of Kelvin probe(*), cable3 and scope

0.2 0.4 0.6 0.8 1

37

36

35

34

0.2 0.4 0.6 0.8 1

10

8

6

4

2

2

4

0.2 0.4 0.6 0.8 11000

1250

1500

1750

2000

2250

2500

0.2 0.4 0.6 0.8 1

60

50

40

30

20

10

50:1 (nominal)

|TF| (dB)

TF (deg)

excl. delay of 8.3 ns

|Z| (Ω)

Z (deg)

freq(GHz)

freq(GHz)

freq(GHz)

freq(GHz)

transfer functionx3m / vD

impedance

2k5 (nominal)

resonanceeffect

(*) GGB – Model 10 - 2k5

© NMDG 2012 15

Primary calibration of TLP systemscope

cable1 cable2

knownDUT

vD & iD

source open short match power meter

source

scope

cable1 cable2

knownactiveDUT

x1m x2mvD & iD

pulse gen

50Ωtermination

x1m x2m

in case of on-wafer the absolute calibration is performed at an “Auxiliary” coaxial calibration plane

© NMDG 2012 16

Additional calibration of TLP system

scope

cable1 cable2

knownDUT

vD

source open

x3m cable3

Compare x3m to calibrated coupler-based vD while measuring an open Transfer function: TF = 1 / e33 = x3m / vD Measure impedance of Kelvin probe during calibration (using iD) Allows to take effect of finite impedance into account during simulation

Kelvinprobe

© NMDG 2012 17

Acknowledgements and References

NMDG wants to thank Rohde & Schwarz and OnSemi for their support

"A new time domain waveform measurement setup to investigate internal node voltages in MMICs",T. Reveyrand, A. Mallet, F. Gizard, L. Lapierre, J.M. Nébus, M. Vanden BosscheMicrowave Technology and Technique Workshop, ESTEC, Noordwijk , May 2004

"Calibrated Waveform Measurement With High-Impedance Probes",Pavel Kabos, Howard Charles Reader, Uwe Arz and Dylan F. WilliamsIEEE Transactions on Microwave Theory and Techniques,Vol. 51, No. 2, pp. 530 - 535, February 2003

"Multiport Investigation of the Coupling of High-Impedance Probes",P. Kabos, U. Arz and D. F. WilliamsIEEE Microwave and Wireless Components Letters,Vol. 14, No. 11, pp. 510 - 512, November 2004

© NMDG 2012 18

Contact info

Want to try this capability?Contact us

atinfo@nmdg.be