IPC-TM-650 EN 2022 试验方法--.pdf - 第472页
T ab le 4-I Resolution of TDR Systems TDR System Risetime Resolution 4X Resolution Figure 4-2 Potential TDR Step A berrations overshoot undershoot ringing low frequency drift IPC-TM-650 Page 4 of 23 Number 2.5.5.7 Subjec…
Figure 4-1 Resolution and Electrical Length of Transmission Line
t
V
adequate resolution
t
V
inadequate resolution
2 T
p
transmission line
IPC-TM-650
Page 3 of 23
Number
2.5.5.7
Subject
Characteristic
Impedance
of
Lines
on
Printed
Boards
by
TDR
Date
03/04
Revision
A
1.1
Measurement
System
Requirements
4.1.1
Measurement
Accuracy
The
measurement
accu¬
racy
of
the
TDR
should
be
sufficient
to
provide
the
required
accuracy
in
the
value
of
characteristic
impedance.
The
required
measurement
accuracy
of
the
TDR
unit
will
depend
on
the
TDR
measurement
method.
In
general,
the
measure¬
ment
accuracy
of
the
TDR
unit
should
be
better
than
1
%
of
amplitude
(either
voltage
or
reflection
coefficient).
Noise
in
the
measured
values
will
affect
the
uncertainty
in
the
calculated
Zo
values.
The
value
of
Zo
may
be
affected
by
the
length
of
the
transmission
line
under
test
and
the
section
of
the
transmis¬
sion
line
from
which
Zo
is
calculated
(see
3.1.1.d).
4.1.2
Temporal/Spatial
Resolution
The
resolution
limit
of
a
given
TDR
unit
is
defined
as
that
particular
time
or
distance
wherein
two
discontinuities
or
changes
on
the
transmission
line
being
measured,
that
would
normally
be
individually
dis¬
cernable,
begin
to
merge
together
because
of
limited
TDR
system
bandwidth.
The
resolution
limit
is
specified
in
either
time
or
distance,
and
is
always
related
to
the
one-way
propa¬
gation
time
between
the
two
discontinuities,
TP
(see
Figure
4-1),
and
not
the
round
trip
propagation
time.
Per
this
definition,
the
resolution
limit
is:
a.
half
the
system
risetime,
tsys,
where
%ys
is
the
10
%
to
90
%
risetime
or
90
%
to
10
%
falltime
(depending
on
whether
the
TDR
response
is
calibrated
with
a
short
or
open
circuit),
or
b.
0.5
kys
x
%,
where
vp
is
the
signal
propagation
velocity
in
the
transmission
line
being
measured.
These
definitions
are
complementary.
For
a
given
length
of
transmission
line
to
be
measured,
the
resolution
should
not
exceed
one
fourth
(0.25)
of
the
available
length,
Ltl
of
the
transmission
line.
Table
4-I
provides
examples
of
required
resolution
for
typical
surface
microstrips
in
air,
and
on
FR4
circuit
board
(%
=
2x1
08
m/s),
for
a
given
TDR
system
risetime.
IPC-2257a-4-1
Table 4-I Resolution of TDR Systems
TDR
System
Risetime Resolution 4X Resolution
Figure 4-2 Potential TDR Step Aberrations
overshoot
undershoot
ringing
low frequency drift
IPC-TM-650
Page 4 of 23
Number
2.5.5.7
Subject
Characteristic
Impedance
of
Lines
on
Printed
Boards
by
TDR
Date
03/04
Revision
A
10
ps
5
ps
/
1
mm
[0.04
in]
4
mm
[0.16
in]
20
ps
1
0
ps
/
2
mm
[0.08
in]
8
mm
[0.31
in]
30
ps
1
5
ps
/
3
mm
[0.12
in]
12
mm
[0.47
in]
100
ps
50
ps/
1
0
mm
[0.39
in]
40
mm
[1
.57
in]
200
ps
1
00
ps
/
20
mm
[0.79
in]
80
mm
[3.15
in]
500
ps
250
ps
/
50
mm
[1
.97in]
200
mm
[7.87
in]
Intermediate
values
can
be
linearly
interpolated
from
Table
4-1
or
using:
法
-三0
For
example,
if
a
32
mm
[1
.26
in]
long
transmission
line
was
being
measured,
a
TDR
system
with
tsys
80
ps
should
be
used.
Note
that,
if
the
probe
launch
caused
excessive
ringing
in
the
TDR
waveform,
or
if
the
launch
does
not
repeatably
replicate
the
connection
to
the
standard,
then
the
0.25
factor
may
need
to
be
smaller.
4.2
TDR
Requirements
4.2.1
Impedance
The
impedance
of
the
TDR
unit
should
be
50
Q
with
an
impedance
uncertainty
less
than
or
equal
to
土
0.5
Q.
This
TDR
impedance
value
is
selected
because
it
is
the
impedance
used
by
most
high-speed/high-frequency
test
instrumentation
and
compatibility
with
this
instrumentation
is
necessary
for
characterizing
the
dynamic
TDR
properties,
such
as
its
impulse
response
(or
transfer
function).
The
imped¬
ance
of
the
TDR
unit
should
be
calibrated
using
an
artifact
standard,
such
as
an
air
line
(see
4.3.6).
However,
the
TDR
impedance
is
a
function
of
frequency
and
calibration
using
a
fixed
region
of
the
TDR
waveform
(the
measurement
zone)
will
only
yield
an
average
impedance
value
for
the
TDR
unit
for
the
corresponding
frequency
range.
4.2.2
Timebase
Accuracy
The
horizontal
timebase
accu¬
racy
defines
how
well
the
TDR
instrument's
horizontal
time
scale
can
display
the
correct
length
of
the
trace.
This
affects
both
the
accuracy
of
the
measurement
zone
calculations
and
any
propagation
delay
values.
The
timebase
accuracy
should
be
less
than
0.25
tsys
(see
also
4.1.2).
4.2.3
Step
Aberrations
The
ability
of
the
TDR
instrument
to
measure
the
impedance
of
a
transmission
line
is
related
to
how
well
the
instrument
can
generate
a
step-pulse
with
a
minimum
of
aberrations
(ringing,
overshoot,
undershoot,
set¬
tling,
etc.).
Any
ringing,
overshoots,
or
undershoots
will
cause
corresponding
aberrations
in
the
TDR
waveform
(see
Figure
4-2).
These
aberrations
can
cause
significant
errors
in
the
impedance
value
computed
from
the
TDR
waveform.
Addi¬
tionally,
low
frequency
step
aberrations
may
produce
a
ramp
in
measurement
zone.
This
ramp
can
cause
a
significant
bias
in
the
computed
impedance
value.
The
TDR
instruments
step
aberrations
should
be
less
than
1
%
of
the
total
step
ampli¬
tude.
For
example,
the
impedance
error
shown
in
Table
4-2
is
for
a
1
mV
error
of
a
250
mV
step.
Poor
settling
and
large
Table 4-2 Impedance Error
Impedance Error
Transmission Line
Impedance
Table 4-3 Connector Torque Specifications
Connector Type Required Torque
Table 4-4 Maximum Suggested Cable Lengths
TDR Cable Assembly TDR Cable Length
IPC-TM-650
Page 5 of 23
Number
2.5.5.7
Subject
Characteristic
Impedance
of
Lines
on
Printed
Boards
by
TDR
Date
03/04
Revision
A
aberrations
will
increase
the
variation
in
the
average
voltage
or
reflection
coefficient
value
from
which
Zo
is
computed,
thereby
increasing
the
variation
in
the
compute
value
of
Zo.
0.24
Q
28
Q
0.4
Q
50
Q
0.79
Q
90
Q
1.23
Q
125
Q
4.3
Other
Equipment
Requirements
4.3.1
Connectors
TDR
systems
typically
come
with
either
“SMA,”
3.5
mm,
or
2.92
mm
connectors
at
their
measure¬
ment
ports.
SMA,
3.5
mm,
and
2.92
mm
connectors
are
all
50
C
connectors
and
are
electrically
and
geometrically
com¬
patible,
therefore,
they
can
be
mated
directly
to
each
other.
However,
the
2.92
mm
and
3.5
mm
connectors
are
precision
connectors
(have
a
lower
impedance
uncertainty
than
the
SMA)
and
are
designed
to
provide
a
more
repeatable
connec¬
tion
than
the
SMA
connector.
Therefore,
for
accurate
mea¬
surements,
it
is
recommended
that
the
2.92
mm
or
3.5
mm
connector
be
used
where
possible.
The
bandwidth
of
the
connectors
must
be
great
enough
so
that
the
connectors
do
not
affect
the
accuracy
of
the
TDR
measurement.
The
typical
-3
dB
bandwidth
of
3.5
mm
connectors
is
approximately
34
GHz
and
of
SMA
connector
is
approximately
24
GHz.
The
reflection
and
insertion
losses
of
the
connector
should
be
less
than
27
dB
and
0.3
dB
respectively.
Other
connectors,
com¬
parable
in
performance
to
the
3.5
mm
connector,
may
also
be
used.
All
cable
connections
using
SMA,
3.5
mm,
or
2.92
mm
connectors
should
be
tightened
with
a
torque
wrench
to
fol¬
lowing
specification,
unless
otherwise
specified
by
the
manu¬
facturer
of
the
connector
or
cable:
[Conversion
factor
is
0.1
128
N-m/(lb-in)]
SMA
0.56
N-m
[5
Ib-in]
3.5
mm
2.92
mm,
K
0.9
N-m
[8
Ib-in]
4.3.2
Cabling
All
test
cables
should
be
coaxial
and
have
a
characteristic
impedance
of
50
Q
with
an
impedance
uncer¬
tainty
of
less
than
±
1
Q.
Cables
used
in
the
measurement
circuit
of
the
transmission
line
under
test
should
have
connec¬
tors
that
are
compatible
with
the
rest
of
the
measurement
system.
The
bandwidth
of
the
cable
must
be
great
enough
so
that
the
cable
does
not
affect
the
accuracy
of
the
TDR
mea¬
surement.
The
length
of
the
cables
should
be
kept
to
a
mini¬
mum.
The
total
insertion
loss
(including
connector
loss)
of
the
cabling
connecting
the
transmission
line
under
test
to
the
TDR
should
be
kept
to
a
minimum,
for
example,
less
than
0.033
dB/cm
[1
db/foot]
at
26.5
GHz.
Table
4-4
contains
sug¬
gested
maximum
cable
lengths.
Faulty
cables
can
contribute
up
to
a
1
Q
error.
Cable
connections
should
be
tightened
with
a
torque
wrench
to
ensure
a
good
connection.
Sampling
Unit
to
Static
Isolation
Unit
30
cm
[12
in]
Static
Isolation
Unit
to
In-Line
Secondary
Standard
91
cm
[36
in]
In-Line
Standard
(such
as,
semi-rigid
coaxial
cable)
1
0
cm
[4
in]
4.3.3
Probes
The
probe
assembly
should
have
a
charac¬
teristic
impedance
of
50
Q
or
of
approximately
the
same
value
as
that
of
the
transmission
line
under
test,
with
an
uncertainty
of
±1.0
Q
or
less.
The
probe
tips
should
be
of
sufficient
diam¬
eter
and
pitch
(spacing
between
signal
and
ground
tips)
to
accurately
and
repeatably
probe
the
desired
probe
contact
pad
geometry.
(See
IPC-2141
for
recommended
probe
land¬
ing
layouts
for
TDR
coupons.)
Single-ended
probes
should
contain
two
tips,
one
each
for
the
signal
and
ground
lines.
Differential
probes
should
contain
two
tips
for
contacting
the
signal
lines
and
one
or
two
tips
for
contacting
the
reference
plane
or
planes.
The
probe
tips
should
have
moderately
sharp
edges
to
cut
through
any
oxides.
For
hand
held
probe
assem¬
blies,
the
probe
handle
should
be
ergonomically
shaped.
The
probe
bandwidth
should
be
sufficient
for
the
desired
temporal/spatial
resolution
(see
4.1
.2).
The
probe
settling
time
should
be
short
so
as
not
to
affect
the
duration
of
the
mea¬
surement
zone.
The
overall
performance
of
the
probe
can
be
incorporated
into
the
TDR
system
response
for
computing
TDR
system
temporal/spatial
resolution
(see
4.1.2).
Inconsis¬
tent
probe
force
and
placement
is
common
and
can
cause
a
significant
yet
unknown
error
that
can
exceed
5
Q.
Probe
connections
should
be
tightened
with
a
torque
wrench
to
ensure
a
good
connection.
4.3.3.
1
Probes
for
Differential
Structures
The
differen¬
tial
probe
should
be
long
enough
to
act
as
a
transfer
stan¬
dard,
similar
to
that
described
in
4.3.7
for
testing
single-ended