to mention the HP8601A shown in Figure 4. This is a
sweep generator that thinks it is an RF generator. Although
most sweep generators don’t really have or need the
accuracy, stability, or precision attenuators that one would
expect in a strictly CW (continuous wave) signal generator,
this one does. It has excellent accuracy, stability, and a
large range of attenuator settings. There are two band
ranges of 100 kHz to 11 MHz, and 1 MHz to 110 MHz.
The sweep function is super linear. If this one is
acceptable for your upper frequency limit, check it out.
Current winning bids on these are $100 - $225.
Next up is frequency counters. No particular
preference here, as there are scads of them out there. You
can’t go wrong with HP, General Radio, and other big
names. There is probably a counter made to suit any
purpose, and if you opt for a universal counter you will
gain period counting along with some other neat
functions. (Even the worst of counters would be suitable
for projects that do not require high accuracy, resolution,
or frequency.)
Here, you want to determine your needs as to RF
work, audio work, etc., and how far you want to resolve
the frequency of interest. I would look for at least a seven-digit display with decent sensitivity and have it extend to
at least 100 MHz with no worse than 10 PPM accuracy. I
prefer LED readouts, but I use both that and the LCD
readout in my lab. Prices are all over the place ranging
from $30 - $400, depending on the hobbyist’s level of
work.
Now for the “King of the Hill” — oscilloscopes! My
clear cut choice here is the Tektronix 2465 series shown
in Figure 5, with bandwidths ranging from 300 MHz to
400 MHz. Also, there’s the 2445 series with the same
features, but a bandwidth of 150 MHz is a nice scope if
the lower bandwidth fits your needs. This may be the last
of Tektronix’ pure analog scopes that have a touch of
digital features included, such as cursors to indicate period
and amplitude along with the current function settings
printed out on the screen. It’s very quick to set up and
operate, with good sensitivity and a wide bandwidth. It
also features two 50 ohm input channels for accurate
readings at high frequencies, and is a fairly small footprint
for a CRT display scope.
There is one caveat here. There’s a history of the X
axis (horizontal) sweep driver chip failing. This chip was a
special run just for this series, and is now unobtainable.
However, this defect may only occur in one out of every
thousand units. Expect to pay $200 - $500 for the 2465,
and $100 - $250 for the 2445, again, depending on the
condition. I reluctantly paid $500 for mine but it had been
recently calibrated and has performed superbly for me.
Analog scopes have what is known as a Gaussian roll-off in regards to the frequency response curve. This is a
gradual and steady drop in display amplitude vs.
frequency, occurring a little before the corner frequency
( 3 dB point). Even though this drop-off occurs, I calibrated
my scope on every range and both channels, all the way
to somewhat beyond 500 MHz. Now, I just use lookup
tables for any frequency displayed, and use a simple
correction factor to know exactly what the true amplitude
is to the displayed level.
In general, analog scopes will have a perfectly flat
amplitude response for about one third of their specced 3
dB bandwidth. This one is no exception, and actually is
perfectly flat out to about 120 MHz. I only mention this as
something to keep in mind for any scope you decide to
purchase.
I have not mentioned the later model flat screen
digitized scopes because they can get pretty pricey,
although Rigol and Tektronix have recently come out with
some low-end models that sell in the $500 range. These
scopes have a pretty flat frequency response within their
rating, but follow a different response curve than the
28 October 2015
FIGURE 4. HP8601A.
FIGURE 5. Tektronix 2465.
