View Single Post
  #5   Report Post  
Old September 10th 09, 03:14 AM posted to
Richard Knoppow Richard Knoppow is offline
external usenet poster
First recorded activity by RadioBanter: Oct 2006
Posts: 527
Default SX-28 image problems

"petev" wrote in message
Thanks for your responses! This SX-28 is far from stock
and is
extremely over-sensitive. I bought it about a year ago to
help me
compare wiring and performance to my 1st SX-28, but this
one was so
modded up, it was no help. The radio works ok, though, but
requires a
of knob tweaking to make a signal sound good. I have to
run the RF
gain at
4 or below. I used to think that was a good thing
until noticing all the cross-mod on 40 and 49 meters.
There is a 6SG7 where the 6AB7 1st RF ought to be.
This receiver also has a 6H6 installed where T5 used to
be.(far rear
There is a 2nd 6H6 installed next to T6 (where the stock
6AB7 used to
So thats 2 6H6s! There is an additional switch on the
faceplate for
the noise limiter,
and the original noise limiter pot was replaced with a
There is also a 6SN7 where the stock 6H6 used to be.
The 6B8 that was with T1(1st IF) has been replaced with a
(The 6B8 pentode wire in T1 has been removed.)
The S meter was replaced with one labeled "carrier" and
has very fast
It does OK on SSB, but the BFO sometimes has a little
Its been re-capped, but not the RF deck. The dial
calibration is dead-
If you have any thoughts as to what these mods were trying
I'm all ears. In the meantime, I'm going to replace the
hotter tubes
with the stock originals,
one by one, and see how that affects performance. I really
intending to do a
complete restore on this SX-28, having done one already,
and have had
that thrill.
Thanks for your help!

There should not be either a 6SG7 or 6AB7 in the
receiver. Both RF stages are 6SK7s and that may be part of
the problem.
When the SX-28 was designed it was equipped with a Lamb
noise blanker. This was a complex circuit intended to blank
out the IF when a noise pulse came along. Such blankers are
fairly common now but its characteristics were not very well
understood at the time so the SX-28 blanker did not work
well. Hallicrafters later offered a modification which
removed the Lamb circuit and replaced it with a stadard
noise clipper. That may account for the extra tube.
Also check to see what was done to the AVC. If its been
modified it may be worth unmodifying it. Note that the use
of high gain tubes may upset the AVC permitting the lower
gain tubes to be overloaded because the high gain tubes will
shut down with much lower AVC voltage than the other tubes
need to be linear.
The SX-28 does not have very good image rejection and
tends to have a lot of spurious responsed due, probably, to
lack of adequate shielding. Image rejection depends on the
bandwidth of the RF stages. Some receivers simply have
better RF than others. For instance, the Hammarlund
Super-Pro, which also has two RF stages and a 455Khz IF has
significantly better rejection than the SX-28 and is pretty
much free of spurs. It also cost nearly twice as much when
I think the noise blanker mod for the SX-28 is included
in the BAMA stuff, I was able to download it somewhere.
There are also differences between the SX-28 and SX-28A,
again, both manuals are available at BAMA and are worth
looking over.
I rather think that Hallicrafters suffered from wishful
thinking in some of their designs. Their aim was usually to
make economically priced receivers of good performance but
they were seldom, if ever, top of the class.

Richard Knoppow
Los Angeles