 |
SIDEBAND
SUPPRESSION ISSUES:
Central Electronics 100/200V
1.0 Introduction
The Central Electronics 100 and 200V exciters utilize the
phase-shift method of single-sideband generation. Early
phasing type exciters, although relatively inexpensive to
construct, offered inferior sideband rejection as compared
to competitive filter-type units. Typically the early sets,
i.e. Central 10A, 20A, etc. used a simple Dome phase shifter
and achieved suppression on the order of only 35 to 40dB.
Additionally, component drift in the phase shifted audio
channels caused relative audio levels to vary, and so, resulted
in a deterioration of sideband suppression and the need
for frequent servicing.
The 100/200V exciters perfected the phase shift method
to a degree that exceeded the capabilities of filter-based
exciters. A special audio phase shifter was developed, using
matched precision components, whose accuracy allowed suppression
in the range of 55 to 60dB. Audio feedback was used throughout
the phase shifted audio channels to make the relative audio
amplitude less sensitive to component aging.
Equally important is the accuracy of the RF phase shifter
used to derive 90-degree offset balanced modulator carrier
injection. The 100/200V use a low-Q, RL/LR network to achieve
the desired 90-degree shift. The inductors of both portions
of this network are slug-tuned to allow for accurate phase
shift adjustment. The low Q of this network minimizes the
possibility of sideband deterioration due to network component
aging.
2.0 Sideband Suppression Adjustment
Before proceeding with the single sideband adjustments,
the technician should familiarize himself with the illustrations
contained in Appendix I. The ultimate objective in this
adjustment process is to obtain an oscilloscope pattern
containing a minimum amount of ripple when a pure sine wave
is applied to the 100/200V audio input. That is, in a correctly
adjusted single-sideband transmitter, a single audio tone
should produce a single RF carrier output (within the measurement
capabilities of a typical oscilloscope presentation).
A low distortion audio oscillator, having less than 0.3%
distortion, is required for the alignment procedure. Oscillators
having distortion higher than 0.3% will introduce distortion
products which would appear as ripple in the oscilloscope
presentation and would limit the ultimate sideband adjustment
to that of the oscillator rather than the exciter under
test.
The following adjustments determine the sideband suppression:
AF Ratio, AF Balance and the 8MHz RL/LR phase-shift network
(total of four adjustments). Before commencing the adjustment
process, verify that the RF Power Output control is set
to maximum level, that the Emission control is set to either
USB or LSB and that the Speech Level control is set to approximately
the 9 O'clock position. The audio oscillator should be set
to approximately 1300Hz and its output level to less than
20 millivolts. The Speech Level control should be adjusted
so that the RF output power developed during the test is
somewhat less than 20 watts. This setting prevents excessive
RF heating and assures that the test signal is below the
clipping level designed for the Audio Limiter stage. Note
that the clipping process of the Audio Limiter stage, if
operational, would introduce high-order distortion products
which could limit the accuracy of the single-sideband adjustment
process.
Adjust the two carrier balance potentiometers for minimum
carrier output as indicated on the oscilloscope. If a noticeable
amount of ripple is present on the oscilloscope, try adjusting
the AF Ratio and AF Balance controls first. Switch between
USB and LSB Emission Switch positions and observe if the
oscilloscope display appears equal. Usually what is observed
is that the ripple is minimum on one sideband setting but
different (worse) for the other. The objective in varying
the two AF controls is to achieve an equal presentation
for both USB/LSB settings. Therefore, try slowly adjusting
the AF controls while frequently changing USB/LSB setting
to achieve the best sideband rejection.
|
|
If adequate suppression is not achieved by
varying the AF Ratio and Balance controls, it will be necessary
to adjust the RF phase shift inductors. Each time these inductors
are varied, it will be necessary to determine that the carrier
is fully balanced out before attempting AF adjustments. It
is often helpful to select the sideband having the worst suppression
and then slowly varying one RF phase shift inductor at a time
to see if there is any improvement. If an improvement is noted,
retune the AF controls for best observed pattern. It may take
several attempts to achieve the best suppression, so be patient!
A much more accurate method for adjusting
these controls requires an RF spectrum analyzer. This device
permits the technician to visually observe the ratio of Upper
vs Lower Sideband suppression, in real time, as the adjustments
are being made. Alternately a communications receiver having
narrow bandwidth filters can be used, in much the same way,
by observing the S-meter differences between USB/LSB Emission
Switch settings.
3.0 Sideband Suppression Problems
Insufficient sideband suppression is a common
problem with newly acquired 100/200Vs. As these sets are nearly
forty years old, it is likely that some corrective action
will be necessary to bring units within specification. It
is recommended that the following be verified prior to any
extensive troubleshooting:
A) Verify that the 6mF/50V Mod Caps are serviceable.
If these are the original white ceramic electrolytics, manufactured
by Chemtronics, they should be replaced by tantalum capacitors
of the same (or somewhat larger) value.
B) Observe that the Carrier Balance potentiometers achieve
balance within their normal range of adjustment. If one is
offset to one end of its rotation, suspect that the 8 MHz
Oscillator plate coil is not correctly adjusted or that one
of the diodes (or socket-pin connection) is defective. If
replacement is needed, a suggested replacement device is the
Motorola MBD701 silicon hot-carrier device. Note that all
four diodes should be replaced whenever one has failed. To
adjust the Oscillator Plate Coil, using an oscilloscope (times
10 probe) measure the signal level present at pin B of T102.
Adjust the coil's slug for maximum 8MHz output level.
C) Verify that the PS-2 audio output levels,
as presented to V-8 (12AT7) grids are equal in level. If measured
levels are found to be unequal, refer to the PS-2 Service
Notes for suggested troubleshooting steps.
D) Inspect the 3900 Ohm 5% carbon composition
resistors feeding the plates of V 8, 9 and 10. On the 100V,
those parts are shown as R146 and 149. On 200Vs the same parts
are shown as R149 and 150. If these parts show signs of temperature
discoloration or measure out of the specified 5% tolerance,
they should be replaced with, minimally, 2 watt 5% tolerance
replacements. Although carbon composition resistors were used
throughout the 100/200Vs, today's metal film resistors offer
superior performance and stability in these audio circuits.
Therefore consider replacement with film types in lieu of
the original carbon types, if available.
E) Check all tubes in the audio stages with
a high quality transconductance tube tester. Replace all weak
or marginally good tubes with high quality replacements.
F) Remember that the sideband alignment process
must be repeated several times to achieve optimal settings.
This takes PATIENCE. But the reward for careful adjustment
is that the exciter will likely require no future retuning,
except in response to catastrophic component failures.
If the difference in ultimate sideband suppression
between USB and LSB settings exceeds 3dB, it is likely that
a small amplitude shift, as a result of component aging, has
occurred in the V-8 through V-10 circuitry. Two cures are
possible. One is to verify that the symmetry of resistor values
in the two audio paths is within tolerances. The other involves
the installation of separate AF Balance potentiometers for
independent adjustment in USB and LSB modes. A modification
kit for this dual AF Balance feature is presently under development.
In the majority of instances, it is possible
to achieve the suppression specification on at least one of
the two sideband modes using the procedures described above.
|
 |
