"geojunkie" wrote in message
om...
Using sweep with the crystal at 60hz is not going to work. Too bad,
because it would be nice to see the pass band shape... I'll have to
check to see if my old sweep generator will take an external input
down at 10hz or lower, but then I don't know how I would sync to my
scope sweep.

The phasing control does have -5 to +5 and there are no stops. 2
sources don't mention where to set it during this alignment and the
third says half meshed and to set the knob to make that zero. It says
to leave it there for the rest of the alignment steps, which includes
peaking the whole IF section to match the crystal, which to me means
1/2 meshed should be on the peak and the null should be off to one
side (either frequency or phase control setting-wise).

I didn't have much luck with AGC voltage, perhaps I was latching on at
the wrong place. I tried the plate of the AGC rectifier and also
downstream of a resistor off the cathode. Didn't see any real swing
with generator frequency anywhere. Perhaps my inexperience really
shows here.

The AVC/AGC usually comes off the detector, so I'm not quite sure what
the AGC rectifier does.

Normally the cathode of the detector is grounded, to minimize hum pickup
from the heater and to give the negative going voltage for the AVC. The
top of the last IF transformer is connected to the diode detector's
plate, and the bottom connects to a RC filter to filter out the IF
frequency. Another RC filter filters out the audio for the AVC voltage.
The AVC bus should be easy to find. One tuned circuit of each gain
controlled stage transformer (IFs and RFs, maybe the converter) connects
to a grid and the AVC bus. The other tuned circuit in each transformer
connects to the plate and B+. If you're lucky the transformer or wiring
will be color coded. B+ will be marked red, the plate is blue, the
grid is green, the grid return (AVC or ground) is black.



So, resorting back to speaker output voltage, I decided to try using
AM from the signal generator with the BFO off. Low and behold, I now
clearly hear and see (on the meter) the twins peaks and intervening
null. I just can't seem to discern them using the BFO as recommended
in all the tech sources. Wonder what I am doing wrong, or why AM
wouldn't be a good way to do this anyway?

The null is sharp enough to discriminate between each of the sidebands
and the carrier. So you can't really null out all of the signal
generator's output as you slowly sweep a modulated signal generator
across the null.

One thing is that I am
injecting post 2nd converter instead of at the tuning stator. Maybe
the BFO would work better as per instructions if I injected at the
stator as per instructions. I noticed that when injecting at the RF
point, slight movements of the tuning knobs changed the peaking
frequency, so I just thought why not inject straight into the IF
section. Am I missing some kind of additional alignment by so doing?


I'm not quite picturing what's happening, and I can't down the schematic
from BAMA to follow along. Anyway, injecting at the stage ahead of the
crystal filter should be fine. Be sure to keep your coupling loose.


Anyway, using AM generator injection into the IF section I now can
tune to the null and hold it pretty steady. I noticed slight tapping
on the crystal bounced the null around, so I opened it up and wiped it
off and retorqued it. Didn't seem to change anything, so mechanical
feedback must be inherent in these beasts.

Yep. The filter is mechanically resonant at the crystal frequency.


Back to that slug on the IF
secondary feeding the crystal filter... the swing between peaks and
null is so vast that on the HP400 I am using they are several scale
settings apart. Makes it hard to tune for maximum swing between
rocking the generator, switching scales, and turning the slug. Is that
for sure what the slug does (maximize the db drop from peak to null)?
Or is it to tune the IF around some certain position of the phasing
control? I was noticing when switching the filter in and out that I
can tune the slug to where the peak without filter is the null when
the filter is switched in (with phasing 1/2 meshed and set to zero)...
is this possibly what I am trying to accomplish or is this just
coincidience?

You shouldn't need to change scales. Try an unmodulated signal and
monitor the AVC line. Be sure you aren't overloading the radio.

Generally, the secondary slug is tuned to the peak. Did you get the
ARRL handbook pages I posted to alt.binaries.pictures.radio? It's a
fair overview of the circuit.



Sorry to be such a nuisance, but I have searched all over the web and
this topic has not been clearly covered anywhere I could find. It
seems worthwhile to reach some kind of understanding as to just how to
align these crystal IF filters. All your feedback so far is really
helping, but I am someone that really wants to understand what the
heck I am doing, and I am not quite there yet. Perhaps this is a great
understatement.

Dan

The tube era ARRL handbooks are worth a look, as well as the internet.
Sometimes it's good to have the book right at your elbow.

Frank Dresser

Just thought I would let everyone know how I ended up doing the
alignment. I used the horizontal (sawtooth) output of the scope (Tek
545a) to drive the external FM input of the sweep generator (Hickok
288x). Set the scope for .1 sec/div (seemed to give about best looking
curve) and the generator for 5kc sweep, then monitored the output with
the scope vertical at volume control. The sweep was feeding the IF
section. Added a straight CW signal from an HP606a to create a "pip"
for calibration purposes, paralled to a 5245L counter.

So what did I see? With the SX-71 set for normal IF (AM signal) I
ended up with a nice bell curve about 4kc wide at the base. When I
switched the crystal filter in I got the sharp spike with adjustable
null as from the ARRL handbook pages Frank posted. Adjusting the slug
seemed basically to be aligning the IF with the crystal. When set for
what to me seemed best, I was able to create an adjustable null (using
phasing control) than ran from just below the cystal peak frequency to
maybe a kc or so below it. When the slug was way off the curve became
very distorted and the crystal would tend to oscillate (ring) much
more. The only unusual artifact I saw was a notch in the higher
frequency side of the peak, not shown in the books. Didn't seem to
hurt anything. It changed with higher sweep rates, so it must be
something due to my test setup.

After setting this slug I aligned the rest of the IF section to the
crystal peak, which was 459khz using just the AM signal from the 606
and monitoring the output on an HP400 rms meter. Everything seems to
be working well, but as this is my first communications receiver I
don't have much to compare it too. The crystal filter is useful on CW,
but also on some SSB signals when an interfering signal causes an
annoying audio signal as it beats with the BFO. I can now tune them
out.

Personally, I think sweep is the way to set this. Then you can really
see what you are doing. Short of that, I am not so sure just finding
the crystal peak and then aligning all the IF slugs to it wouldn't do
the job just fine. Thanks to everyone for their very helpful advice.

Dan