Text preview for : 5989-0270EN.pdf part of HP 5989-0270EN HP Publikacje 5989-0270EN.pdf



Back to : 5989-0270EN.pdf | Home

Non-Zero Noise Figure After Calibration

Application Note 1484

The increasing demand for wireless products combined with the availability
of modern easy-to-use measurement equipment mean that noise figure
measurements are more popular than ever. This is especially true in the
microwave region of the spectrum where more and more wireless standards
are being developed.

Noise figure measurements rely on a calibrated noise source as a reference.
The general measurement process is shown in the figures below. In Figure 1, the
noise source is connected directly to the input of the measuring instrument
and a user calibration is performed. This measures and stores the instrument's
own noise figure at its various attenuator settings. These results are used to
remove the effect of "second-stage" noise contribution during a corrected
measurement. The measurement arrangement, with the device-under-test (DUT)
inserted between the noise source and the instrument is shown in Figure 2.



Calibration setup

Noice source




Figure 1. Calibration



Noice source
F1, G1

DUT
F2

Figure 2. Measurement


Immediately after calibration, the noise source is still connected directly to
the instrument, which automatically switches to the corrected measurement
mode. In this configuration the instrument would be expected to display a
noise figure and gain of 0 dB, because there is no DUT present. In practice
however, the instrument may show a noise figure of plus or minus a few
tenths of a dB as well as an even lower level of gain, but not zero. While this
is quite normal, some users assume that this zero-error will be added to their
DUT's noise figure, producing an inaccurate result.

This application note attempts to explain the reasons behind the zero-error
and shows with examples that it does not have a compromising effect on
instrument accuracy in a measurement situation.
Theory
Figure 2 along with the basic noise figure equation will be used as the basis
of the explanation. The general equation for the noise factor of two cascaded
stages as defined by Friis (Proceedings of the IRE, July 1944, pp.419-422) as:

(F2