Specifications

5
Source AM noise suppression
For absolute phase noise measurements, AM
noise contributions to the baseband signal (after
the phase detector) are typically well below the
phase noise contributions and therefore are
usually ignored. For a residual phase noise meas-
urement, however, the contribution to the base-
band signal from AM noise can often exceed the
contribution of the system phase noise contribu-
tion (measurement system phase noise floor).
The largest contributor to AM noise present at
the input ports of the phase detector is the stimu-
lus source used for the residual measurement.
The typical suppression of AM noise is 20 to
30 dB for an RF phase detector and 10 to 20 dB
for a microwave phase detector. As shown in
Figure 2-1, the AM noise of a stimulus source
can significantly influence the measurable noise
floor of a residual phase noise measurement.
Measurement offset range
The offset range for the measurement is limited
to PRF/2, based on sampling theory.
Phase transients
Since the DUT is pulsed, phase transients may
appear at the output of the phase detector. If
these phase transients are large enough, the test
set LNA gain may be reduced, degrading the over-
all measurement noise floor. In the 70420A phase
noise test set, LNA overload is monitored by a
DC-coupled peak detector on the output of the
LNA. The overload setpoint, which has been estab-
lished for CW operation, is set at ±2.5 V
pk
at the
output of the LNA or ±0.5 V
pk
at the monitor port.
As shown in Figure 2-2, the amplitude of the lead-
ing or trailing phase transient may exceed the
CW trip point, but only for a fraction of the pulse
width. Even though the LNA is not experiencing
an overload condition across the entire pulse width,
the system—having sensed an overload condition—
will remove LNA gain from the measurement path.
To overcome this situation, it is acceptable to over-
ride the hardware overload detector.
Figure 2-1. AM noise floor versus Phase noise floor