Keysight Technologies E4412A Operating And Service Manual page 21

Operation and Service Guide 1
The chart in Figure 1-2 shows the typical uncertainty in making relative power
measurements with the E44XX-Series power meter in the Rel (relative) mode. In a
relative measurement, two power levels are compared using a single sensor. The
typical measurement uncertainty can be found by drawing a vertical line at a
power level used as a reference, and a horizontal line at the power being
measured. The region where these two lines intersect shows the typical
uncertainty. This assumes that the reference power and the measured power are
at the same frequency, and neglects errors due to zero set, zero drift, and noise. It
also assumes no change in mismatch when measuring the "Power level used as
reference" and the "Power level being measured". This chart illustrates that the
best relative power measurement accuracy is obtained when the reference and
the measured power levels are equal to, or less than, –20 dBm. Care in choosing
the power levels can yield improved measurement accuracy. Example A illustrates
a relative gain (amplifier measurement) and example B illustrates a relative loss
(insertion loss measurement).
EXAMPLE A (Amplifier Gain Measurement):
Input power (reference) = –35 dBm; power measured = +6 dBm' so the relative
gain is +41 dB. the chart indicates a typical error of ±6 %, which corresponds to
+0.25 dB/–0.27 dB uncertainty in the gain measurement.
EXAMPLE B (Insertion Loss Measurement):
Reference power = –25 dBm; power measured = –35 dBm, for the case of an
insertion loss of 10 dB. the chart indicates a typical error of ±1 %, which
corresponds to ±0.04 dBm uncertainty in the measurement.
Keysight E4412A and E4413A Operating and Service Guide 21