Table of Contents
Advertisement
7
Residual Measurement Fundamentals
4
5
6
Calibration options
There are five calibration methods that to choose from for calibrating a
two-port measurement. The procedure for each method is provided on the
following pages. The advantages and disadvantages of each method are also
provided to help you select the best method for your application. The primary
considerations for selecting a calibration method are:
• Measurement accuracy
• Equipment availability
208
DUT and other phase-sensitive components from mechanically-induced
phase noise. The mechanical shock of bumping the test set or kicking the
table will often knock a sensitive residual phase noise measurement out of
quadrature.
When making an extremely sensitive measurement it is essential to use
semi-rigid cable between the components. The bending of a flexible cable
from vibrations and temperature variations in the room can cause enough
phase noise in flexible connecting cables to destroy the accuracy of a
sensitive measurement. The connectors also must be tight; a torque wrench
is the best tool.
When measuring a low-noise device, it is important that the source and any
amplification, required to achieve the proper power at the phase detector,
be placed before the splitter so it will be correlated out of the measurement.
In cases where this is not possible; remember that any noise source, such as
an amplifier, placed after the splitter in either phase detector path, will
contribute to the measured noise.
An amplifier must be used in cases where the signal level out of the DUT is
too small to drive the phase detector, or the drive level is inadequate to
provide a low enough system noise floor. In this case the amplifier should
have the following characteristics:
•
It should have the lowest possible noise figure, and the greatest possible
dynamic range.
•
The signal level must be kept as high as possible at all points in the setup
to minimize degradation from the thermal noise floor.
•
It should have only enough gain to provide the required signal levels.
Excess gain leads to amplifiers operating in gain compression, making
them very vulnerable to multiplicative noise problems. The non-linearity
of the active device produces mixing which multiplies the baseband
noise of the active device and power supply noise around the carrier.
•
The amplifier's sensitivity to power supply noise and the power supply
noise itself must both be minimized.
Agilent E5505A User's Guide
Advertisement
Table of Contents
