HP 8753D - Page 212

T
able 5-2.
Purpose and
Use of
Dierent
Error-Correction
Procedures
Correction
Procedure
Corresponding
Measurement
Errors
Corrected
Standard
Devices
Response Transmission
or reection
measurement
when
the
highest
accuracy is
not required.
Frequency
response.
Thru
for transmission,
open
or
short
for
reection.
Response
&
isolation
Transmission
of
high
insertion
loss
devices or
reection of
high return
loss
devices
.
Not
as
accurate
as
1-port
or
2-port
correction.
Frequency
response
plus
isolation in
transmission or
directivity
in
reection.
Same
as
response
plus
isolation standard.
(load)
S
11
1-port Reection
of
any
one-port
device
or
well
terminated
two-port
device.
Directivity
,
source
match,
frequency
response
.
Short
and
open
and
load.
S
22
1-port Reection
of
any
one-port
device
or well
terminated two-port
device
.
Directivity
,
source
match,
frequency response
.
Short
and
open
and
load.
Full 2-port Transmission or
reection of
highest
accuracy
for
two-port
devices
.
Directivity,
source match,
load
match,
isolation,
frequency
response
,
forward
and
reverse
.
Short and
open and
load
and
thru.
(2
loads
for
isolation)
TRL
*
/LRM
*
Transmission
or
reection
when
highest
accuracy
is
not
required.
Directivity
,
isolation,
frequency
response
.
(forward and
reverse)
Thru,
reect,
line
,
or
line
,
reect,
match,
or
thru,
reect, match.
Note
Frequency
response
calibration
is
not
as
accurate
as
other
calibration
methods.
Error-Correction
Stimulus
State
Error-correction
is
only
valid
for
a
specic
stimulus
state,
which you
must select
before you
start
a
correction.
If
you
change
any
of
the
following
parameters,
you will
invalidate the
correction
and
the
analyzer
will
switch
the
correction
o
(unless
the interpolated
error
correction
feature
is
activated):
frequency
range
number of
points
sweep type
The error-correction quality may be degraded (
Cor changes to
C? ), if you change the following
stimulus state parameters:
sweep time
system bandwidth
output power
Note
If you activate averaging, the analyzer may increase the sweep time
if more
time is needed to calculate the averages. If the sweep time changes, you will
see Cor change to C? . The number of averages does not aect a sweep cycle
time. Therefore, if you use averaging for error-correction, leave it on for the
measurement and set the averaging factor to 1, for a faster sweep.
Optimizing Measurement Results 5-5