Video Fi Lter; Sensitivity - Maximum Input Level; Frequency Response; Functional Principle Of The Hm5530 - Hameg HM5530 Manual

S p e c t r u m a n a l y z e r s p e c i f i c a t i o n s
The equation shows that the noise power is directly proportional
to bandwidth. Hence reducing the fi lter bandwidth by a decade
will decrease the noise by 10 dB. This is equivalent to a sensi-
tivity increase by 10 dB.
All other noise sources within the analyzer are assumed to be
non-thermal. Sources of non-thermal noise are: undesired
emissions, distortions due to nonlinear characteristics or
mismatches. The non-thermal noise defi nes the socalled noise
fi gure to which the thermal noise is added in order to arrive at
the total noise fi gure of the system. This is the noise which is
visible on the screen and which determines the sensitivity of
the analyzer.
As the noise level depends on the bandwidth, any comparison
of analyzers requires the use of the same bandwidth and the
same bandwidth defi nition (–3 or –6 dB). Spectrum analyzers are
swept over a wide frequency range, but they are narrow band-
pass selective measuring instruments. All signals within the
frequency range of the analyzer are converted (possibly several
times) to an if (or several) and pass the if fi lter(s). The detector
at the if output sees only that noise which passes through the
narrowest fi lter, and this will be displayed. When measuring
discrete signals, maximum sensitivity is hence achieved with
the narrowest fi lter bandwidth.
Video fi lter
If the amplitude of a signal is comparable to the analyzer's
average noise, a measurement becomes diffi cult. The measure-
ment can be improved by reducing the bandwidth below that
of the narrowest if fi lter. A socalled video fi lter is inserted in
the signal path following the detector, its bandwidth of 4 kHz
averages the instrument's noise and decreases the displayed
noise substantially. In many cases a small signal buried in noise
will become visible.
If the if bandwidth is already small compared to the span se-
lected (high sweep speed), the video fi lter should not be used,
because this could lead to false (too low) amplitude measure-
ments. (An illegal combination of fi lter bandwidth and sweep
speed will be indicated by „uncal" in place of the sweep time
readout (SW ... )).
Sensitivity – Maximum input level
The defi nition resp. specifi cation of an analyzer's sensitivity is
somewhat arbitrary. One method of specifi cation is to defi ne the
sensitivity as that input signal power level which is equal to the
analyzer's average noise power level. As an analyzer measures
signal plus noise, the signal will appear 3 dB above the noise.
The maximum permissible input level is that which is still
safe for the input stage. This level is specifi ed as + 10 dBm (no
attenuation, attenuator 0 dB) and + 20 dBm (attenuator 10 to
50 dBm) for the input mixer. Before the „burn-out" level is
reached, the analyzer will start to compress the signal; this is
acceptable as long as the compression remains below 1 dB.
The analyzer will also produce nonlinearities if overdriven. There
is further the danger of undetected input stage overload because
individual spectral lines may only change imperceptibly due to
the onset of compression. In such cases the amplitude display
will not any more be true.
The analyzer generates distortions, mostly by input stage non-
linearities. These remain 75 dBc below the input signal level
>
as long as the level is –30 dBm.
<
38
Subject to change without notice
Larger input signals should be reduced by the attenuator prece-
ding the mixer. The largest signal which the analyzer can digest
without creating more distortions than specifi ed is called the
„optimum input level", meaning that the mixer input remains
–30 dBm. At higher levels, the specifi cation for the generation
<
of harmonics will not be met any more. The distortionfree input
range is also called the „useful dynamic range". This is to be dif-
ferentiated from the display range which is the ratio of the highest
to the lowest signal displayed simultaneously without any visible
intermodulation products.
The maximum dynamic range follows from the specifi cations.
The fi rst hint is the specifi cation for the harmonics' level, this is
75 dBc below the signal as long as the input level to the mixer
>
is –30 dBm. In order to make full use of these specifi cations,
<
the analyzer must be able to detect levels of –110 dBm. The if
bandwidth required for this should not be too narrow, otherwise
diffi cuties will arise due to noise sidebands.The if bandwidth of
9 kHz is suffi cient to display signals at this level.
The distortionfree measuring range may be further extended
by reducing the input level. This is limited by the analyzer's
sensitivity. The maximum available dynamic range is achieved if
the highest peak of the spectral lines just touches the reference
level. i.e. the top of the graticule.

Frequency response.

The frequency response should be fl at over the range, i.e. the
accuracy of the signal level measured should be independent of
frequency. Amplifi ers and fi lters must be given suffi cient time
to reach full amplitude.

Functional principle of the HM5530

The HM5530 is a spectrum analyzer for the frequency range of
100 kHz to 3 GHz. The spectral components of signals in this
range can be detected and measured from –110 to +20 dBm.
The signal to be analyzed fi rst passes through an attenuator
which can be switched from 0 to 50 dB in 10 dB steps. A pre-
selection input fi lter follows which serves several purposes: to
some degree, it prevents multiple signal reception, it prevents
the reception of signals at the 1st if (if feedthrough), and it sup-
presses any oscillator feedback to the input. The purpose of the
input mixer and the 1st oscillator (1st LO) is the conversion of the
input frequencies within the analyzer's range; it determines the
frequency dependent amplitude characteristic and the dynamic
properties of the instrument.
The analyzer is designed as a triple superheterodyne receiver,
it is an electronically tuned selective amplifi er. Frequency tu-
ning is performed with the aid of the 1
through the range of 3537.3 to 6537.3 MHz. Its output signal
and the full-range input signal are fed to the fi rst mixer (input
mixer). At the mixer output there are the following frequency
components present:
st
1. Signal of the 1 LO, the frequency of which must be 3537.3
MHz above the frequency of the desired input signal. The
st
frequency of the 1 LO will thus be 3537.3 MHz if the input
st
LO which can be tuned