A Step by Step Guide to Practical RF Device Measurement

7/29/2019 A Step by Step Guide to Practical RF Device Measurement

1/33

&RPSDQ\&RQILGHQWLDO 6HHLW7RXFKLW0HDVXUHLW

$VWHSE\VWHSJXLGHWRSUDFWLFDO5)GHYLFH

PHDVXUHPHQWGavin Fisher

Cascade Microtech Europe


2/33


,QWURGXFWLRQ

Device measurement at its simplest requires the following

steps

Probe and accessory physical Set-up for calibration

Calibration instrumentation setting preparation

Calibration raw measurement steps, error computation andcalset delivery to instrument

Calibration verification

Wafer alignment and test plan set-up Device measurement, including biasing

Measurement of de-embedding structures and post process

activities to remove unwanted parasitics


3/33


7\SLFDO6\VWHP

Vector Network

Analyzer

Cables

Probes

Probe positioners

Probe station

Contact Substrate

Calibration Substrate

Calibration Software


4/33


Dry, Frost Freeenvironment

Auxiliary Chucks Roll-out chuck

Stable repeatableplaten

Top-Hat

0LFUR&KDPEHU70 7HFKQRORJ\


5/33


3UREH3ODQDUL]DWLRQ

*definition: planarization -

the ability to ensure all

contacts are at the same

height

Use Contact

Substrate to

planarise probes

Equalprobemarks


6/33


3ODQDUL]LQJ WKH3UREHV

Contact Substrate

PN 005-018

Dull gold finish

Bright contact marks

Adjust planarity until

equal marks from all

probe contacts


7/33


5),PSHGDQFH6WDQGDUG6XEVWUDWHV

Electrical behavior of standards

Standard dimensions Probe pitch

Probe placement (Use Alignment marks)

Cal coefficients supplied with probe & ISS


8/33


,66$OLJQPHQW0DUNV

Sets probes overtravel & spacing

for calibration

Initial Contact (zero overtravel)

Line the edges of the probes to

edge of flags

Center the contacts with X & Ymicrometers

Final Contact (2 3 mils overtravel)

Tips lined up with flag centers

Center the contacts with Z

micrometer only


9/33


3UREHDOLJQPHQWRQ,VV


10/33


:LQ&DO;(

Tools for the novice

Guided Wizards

Multi-media Tutorials

Intelligence in setups

Tools for expert Enhanced verification

Real time measurement

validation Enhanced reports


11/33


6\VWHP6HWXS

Measurement System Setup

Define the measurement system

VNA, prober, ISS and probes

VNA Qualification

Test that the VNA is functional

and repeatable

Probe Qualification

Check that the probe is making

contact

ISS management

What structures to use

Is a structure good?


12/33


8VLQJ:LQFDO;(WR3UHSDUHWKH

FDOLEUDWLRQ

Important to initialise instrument settings paying attention to power, number of points, Start and stop

and particularly IF bandwidth


13/33


3UREH6HWXS

Probe characteristics are

displayed both graphically

and numerically. Probes can

be identified by serialisation

Probe data required to

check calibration

compatability and where

necessary provide lumped

element data


14/33


,666HWXSIRU$XWRFDOLEUDWLRQ

Individually serialised iss

data can be loaded

This information is

important to keep track of

correct iss for calibration

and determine location of

alignment structure


15/33


,66$OLJQPHQWVWUXFWXUHORFDWLRQ

ISS Reference location

determines the correct orientation

and alignment of the probes withrespect to the entire iss

A similar tool is used to inform

the software of damaged or

untrimmed lcations


16/33


$XWRPDWLF&DOLEUDWLRQ


17/33


&DOLEUDWLRQ3URFHGXUH

Automatic calibration will use the prober to automatically move from

standard to standard

On pressing autocal the procedure is as follows

Repeatability check measures raw open multiple times in order to

check the system is repeatable (often picks up problems relating to

cabling, system directivity, Excessively high If bandwidth) Calibration moves though all standards for the calibration,

computes calibration and sends to instrument

Verification will look at a verification standard to compare against

known values (typically an open)

Monitoring measurement will store data for future checks against

system stability (is cal still good)


18/33


5HSHDWDELOLW\&KHFN

Wincal measures open to check repeatability of measurement system


19/33


&DOLEUDWLRQPHDVXUHPHQWVIRU/5507KUX


20/33


&DOLEUDWLRQPHDVXUHPHQWVIRU/5502SHQ

System re-measures open for the calibration. At times the

open measurement uses substrate opens hence the need

for remeasurement


21/33


&DOLEUDWLRQ0HDVXUHPHQWVIRU/5506KRUW


22/33


&DOLEUDWLRQ0HDVXUHPHQWV /RDG

It is important that only 50 ohm loads are used for this part

of the calibration


23/33


&DOLEUDWLRQ &RPSXWDWLRQDQGVHQGLQJRIHUURUFRHIILFLHQWV

Wincal applies the selected calibration to the measureddata (typically we recommend LRRM) and error set is sent

to the instrument


24/33


&DOLEUDWLRQ 9DOLGDWLRQ

Following calibration a validation is carried out against a known standard. Typically thisis an open whose capacitance is known by the probe pitch, but can be a golden dutwhose characteristics are pre-measured and stored. For lrrm the open is the raw openmeasured during the cal and corrected by the calibration (post corrected)


25/33


&DOLEUDWLRQ 0RQLWRULQJ

Following validation wincal will take a measurement of a

standard for use in later monitoring tests.

Monitoring test allow Wincal to gain a metric of system

drift and thus the continuing validity of the calibration.

An open is an easy standard to reproduce and is affectedlittle by probe movement ie this can often be measured

just above the device under test


26/33


,QGHSHQGHQW9HULILFDWLRQ6WDQGDUGV

Openstub

GROUND

GROUND

SIGNAL

0.2 0.4 0.6 0.8 1 1.5 2 3 4 5 10 2050

0.2

0.4

0.6

0.81

1.5

2

3

4

5

10

20

50

-0.2

-0.4

-0.6

-0.8-1

-1.5

-2

-3

-4

-5

-10

-20

-50

0.1 GHz

40.0 GHz

-1.5

-1.0

-0.5

0.0

0 5 10 15 20 25 30 35 40

[dB]

[GHz]

Linear Phase Lag

50ohm to stub Z miss-match

Fringing C at stub open end


27/33


:DIHU$OLJQPHQW

Prior to carrying out any

multi-die test the wafer

should be aligned to the

system axes and

appropriately sized.

Nucleus can carry out this

activity automatically using

optical pattern recognition

If an exisiting wafermap is

available his can spped up

alignment process


28/33


:DIHUDOLJQPHQW

Using Evue microscope multi ccd view makes location of usable

optical alignment points straightforward


29/33


:DIHUPDSSLQJ

Nucleus wafermap allows test plan to be visible and test progress to

be tracked


30/33


:DIHUPDSSLQJ VXEGLH

Subdie allow for multiple devices to be identified and

tested within each die


31/33


:DIHUPDS SDUDPHWHUPDSSLQJ

Wafermaps can also be used to see the distribution of

single point data for die and subdie as the test progresses

This tool is most useful for spotting set-up issues during

the test sequence


32/33


([DPSOHVRIXVLQJVXEGLH IRUWHVWV

( O I L :L O ;( I


33/33


([DPSOHRIXVLQJ:LQFDO;(IRUVSRWPHDVXUHPHQWVDQGFRPSDULVRQ

Documents

A Step by Step Guide to Practical RF Device Measurement