subbed in a pair of 10-turn pots I
had in stock.
Funny thing is that by the time I
received my three-turn pots, I was
delighted with the way the 10-turn
pots performed. I originally thought
these would have required too
much fiddling around to adjust
them. However, they were really
not that bothersome — especially
when you consider that they are
basically for setting up the sweep
parameters, and then just get on
with the sweep testing at hand. The
resolution is such that it
overshadows any minor
inconvenience of their usage.
Needless to say, the 10-turns
won out and the newly acquired
three-turns were stored away for
use in another project. While we
are on the subject of multi-turn
pots, I have to mention that about a
year ago I replaced the fine-tuning
control on my RF generator with a
five-turn pot and it made a world of difference in tuning to
an exact frequency — especially on the higher bands. You
may want to do the same, as it is well worth the effort.
There are a fair amount of wires interconnecting the
circuit board and front panel. I wanted to bundle them,
yet easily be able to identify them at the other end by
color-coding them. Having only seven or eight colors to
choose, I quickly ran out of options. So, I used a trick from
the past to reuse the same colors by marking them with a
continuous stripe. Clamp one end of the wire in a vice
and while drawing the wire tight, trace a line down its
length with a black felt tip pen. A blunt one works the
best and one pass is adequate. Additional passes will
remove more ink than they lay down.
Several precision metal film resistors are used in this
circuit. Their actual ohmic value is not super critical, and
near values may be used as long as they are within a few
percent of the schematic values. Basically, they are used
for their tighter temperature stability. Resistors R15, R16,
and R17 can be off value somewhat, but the total of their
combined resistance should be close to the design value
of 10K as this is part of the calibration string. After
completing construction, check all controls for operation.
If they are in the ball park, you can proceed with
The original design of my RF generator calls for a Vt
+0.6 VDC to - 6. 2 VDC which is a span of 6. 8 VDC. Due
to component tolerances, this could vary from unit to unit.
For now, let’s assume that they are all identical to establish
a standard of calibration and test setup procedure. I only
mention this if your situation is slightly different and the
calibration voltages might have to be changed. I tried to
keep the calibration pot resistance small in relation to the
overall strings of resistors associated with them. This is due
to the fact that they would be the most unstable resistance
of those networks, and I wanted the 1% metal film
resistors to make up the bulk of the load because of their
superior stability. So, if you run out of pot travel in the
calibration procedures, you will have to go up or down in
value from the associated resistor of that particular string
(i.e., R10, R13, R14, R22).
Begin the calibration procedure by turning the Manual
sweep pot and the Vernier fully clockwise; S4 at the 100%
position. Place S1a in the Auto position, S3a in the Run
position, and the Sweep Rate at mid position. Scope
probe IC1b’s output; you should see a very linear ramp of
about 8V peak. Place the scope probe at the wiper of S4
in the 100% position and adjust the 2K cal pot that feeds
this switch for 6.8V peak.
Now, place S3a to the Set position and adjust the 5K
pot feeding that Set position for a DC voltage exactly the
same level as the sawtooth peak ( 6. 8 VDC). Set S3a back
to the Run position and S1a to the Manual position.
Adjust the 2K cal pot feeding that control for exactly the
same voltage as the previous DC voltage ( 6. 8 VDC).
Now, the Auto sweep level (ramp) and Manual sweep
level (DC) will always track each other exactly, no matter
how the testing setup is arranged. Of course, the scope
trace and Vt ramp are always in sync by virtue of being
locked on the circuit board. Leave these controls set as-is
February 2017 35
■ FIGURE 4.