Input Characteristics - HP 8340A Operating Manual

Model 8340A
Narrow pulse leveling accuracy problems may be treated by simply not attempting to level
them. The unleveled mode allows the user to control the linear modulator directly. The
setting of this modulator does not change with pulse width, so the pulse amplitude remains
constant as the pulse is narrowed. Pulses as narrow as 25 nsec may be produced in this
mode. To set the amplitude, go to CW operation (pulse off) and set the desired power via the
POWER dBm display.
As the pulses are narrowed, the POWER dBm reading drops since it is measuring the output
of the detector S/H system and it is the limitations of this system which cause the narrow
pulse leveling problems. The real amplitude remains essentially constant, however. In
unleveled mode, the POWER dBm display accuracy is the complement of the corre­
sponding leveled pulse accuracy; that is, -1.5 dB
frequency> 2.3 GHz.
drift (a few tenths of a dB) due to component self-heating when going from 100% duty cycle
to low duty cycles.
Since unleveled mode does not involve the integrate/hold circuit, there is no leakage
induced amplitude drift between pulses. Consequently, very long periods may be employed
that are limited only by the long term drift of the modulator drive circuits. This is not
negligible and should be characterized by the user at the frequency of interest. Drifts of a
few dB should be expected during warmup. Since this mode is not feedback leveled, the
power changes markedly with frequency.
In unleveled mode the ALC inputs control the linear modulator directly, so the response
time is not dependent on pulse parameters. The response time is 70 µ,sec maximum in CW,
sweeps > 5 seconds, normal AM; or 4 µ,sec maximum for sweeps < 5 seconds or with
AM works in unleveled mode, but the linearity and sensitivity varies with
[SHtFn[AM].
power level and frequency. See the pertinent preceding section.

Input Characteristics

When pulse mode is activated, the HP 8340A's RF output is controlled by the voltage at the
pulse input. The input circuit is shown in Figure 3-36. The output is off for inputs below
approximately + 1.5 V. If the irrput is left open it pulls itself up to + 1.8 V, which will
output on. The HP 8340A's delay and compression specifications assume an input of 0 to
+3.0 V, which requires sinking about 1.6 ma and sourcing about 2.7 ma respectively. The
input accepts + 12 V to -20 V with no damage, which is compatible with the + 6 V
modulator drive of certain network analyzers. Aside from small effects on delay and
compression, the waveform and amplitude of the input is unimportant, as long as it
transitions from +0.5 to +2.5 V in 2 µ,sec or less. This transition is sufficient to generate the
fastest rise time. Since the input is not linear, input -Overshoot will not appear on the
output.
Some applications may require a 500 input. One simple solution is a 500 feedthrough
termination. In the case of special configurations with rear panel inputs, the long length of
unterminated internal cable may still cause reflection problems. If so, R14 (circuit board
A21) may be replaced with a 500 resistor. The consequences are that the input no longer
pulls up when open, so activating pulse mode with no input will shut off the RF output.
With a 1/4 watt resistor the damage level is+ 3.5 V. A 3/4 watt rating is needed to accept the
+6 V drive from certain network analyzers.
If operating close to maximum ALC output there is some amplitude
Scans by HB9HCA and HB9FSX
Operating Information
+ 0.3 dB for width
=
100 to 200 nsec,
\
turn the
3-111