If you don't have one, you're not going to be able to tune that
filter accurately. You -might- be able to fake it with a generic sweep
generator and a spectrum analyzer, but it'll be tricky because you need
to EXACTLY match the sweep rate of the generator with that of the
spectrum analyzer. If they're not matched, you'll never get an accurate
tune.
Actually, that's what I've been doing so far. Since we don't have a
tracking generator, I've just been using our specan, (hp 8568a) with
our 8660b in fast auto sweep mode, and it just so happens to sync up
with the spec an w/ a sweep time of 100mS (very tricky indeed), and I
can only use this for ballpark adjustments. To tune it in I have to
put it in single slow sweep mode, and put the specan in max hold and
wait for the sweep to finish (takes forever each cycle). Now,
however, our spec an failed its performance test, specifically the
resolution bw switching uncertainty test.
The table is:
all measurements in dB, tol at 3Mhz and 10Hz +-1dB, all else +-.5dB
3Mhz .07 in
300kHz -.28 in
100kHz -.92 out
30kHz .69 out
10kHz .41 in
3kHz .32 in
1kHz 1.84 out
300Hz -.13 in
100Hz -.37 in
30Hz -1.08 out
10Hz -7.70 way out
This is where I stopped the performance tests, I know I should go
through them all before I go adjusting, but it is a loooong procedure,
and if I just have to redo it all again....
Hoping that it is just adjustment, not repair needed,...
P/ the S.M., the related adjustments for this procedure are
3MHz b.w. filter
21.4 B.W filter
Down/Up converter
The 3MHz filter adjustments call out for the tracking generator. To
set up the equipment, verbatim from the manual:
8. Disconnect 97 (white/violet) cable from A4A8J1 and connect
tradcking generator to A4A8J1 using BNC to SMB snap-on cable.
9. Set spectrum analyzer SCAN WIDTH PER DIVISION to 20kHz and red
scan width knob to ZERO. Set TUNING STABILIZER switch to ON. Set
tracking generator output level to -25dBm and tune spectrum analyzer
FREQUENCY for a tracking generator output frequency of 21.4000MHZ.
There is only a single cable from the tracking generator going into
a4a8j1, and there aren't any sync lines between the specan/tracking
generator and the 8568a.
**So is this just a center frequency of 21.4MHz w/ a sweep width of
200kHz, or is it CW at 21.4MHz?
The procedure starts with filter peak adjust, and to connect scope
chA to tp7, chB to tp5, adjust tracking generator output frequency to
peak ch a display, then adjust a cap for maximum peak to peak level on
chB. This is then repeated at different test points, and adjusting
different caps,.. and finally using the spec an crt to achieve final
peak.
Then for center and symmetry. Still nothing about syncing the
tracking generator to the 8568a. It says to bypass, one by one, the
xtal filters and adjust for minimum peaks and symmetry.
This all makes sense, but there isn't any adjusment of sweep time, or
syncing the two spec an's, or anything like that.
The other tests only require DVM, step attenuator, power meter, etc..
So, what would you make of this? Is the tracking generator inputting
a swept frequency to the 8568, or is it CW? Think it's possible to
use the 8660b for this?
If you can tell me which SpecAn you're working with, I may be able
to suggest a matching generator. HP 8568a.
Happy tweaking.
Also, the manual calls out for an HP 141T/8552B/8553B/8443A.
If I need to clairify further, please let me know. Sorry if this is
confusing, and thanks for the quick response.
Thanks, Steve