[HBR] Need help with noise

[Ian Wilson]

There were 3 major problems, two of which had to do with the preselector/RF
amp stage: poor shielding between sections of the dual-gang tuning
capacitor,
and horrible layout around the RF stage (caused by an unforeseen change in
approach to the front end layout). Tackling these has stabilized the RF
stage
nicely. There may well be problems on higher bands so I need to improve
things
more, though. I have a 1k cathode resistor on the 6EH7 so this is not
running
flat out (will be adding a manual gain pot).

Another problem was caused by poor design in the coupling between the 2nd
mixer and the 455kHz mechanical filter (these are some distance apart). Some
circuit simulation work revealed an enormous amount of loss. Fixed now.

As a result there is now rather too much gain to run with no AGC with a few
feet of wire as an antenna. This is what I was expecting; I can modify the
gain
distribution later.

Still lots to do but the basic stuff seems to be working well now; thanks
again
for the help.

73, ian K3IMW



On Sat, Jan 18, 2014 at 6:28 PM, Walt Hutchens <waltah at earthlink.net> wrote:

> Ian said:
>
> > With the RF amp replaced by a 5pF capacitor (turning the two LCs
> > into a sort of double tuned preselector), the receiver behaves way
> > better, with what I regard as normal S/N for a not-very-narrow BW
> > system (the IF filter is 2.5kHz wide).
>
> EXCELLENT!
>
> The blind alleys I've slept in myself are easier to recognize than the
> others ...
>
> > My intention for the RF amplifier was to provide a small amount of
> > gain (for poor antennas) with a manual RF gain control to act as an
> > attenuator for strong signals. I expect the 6EJ7 to have good intermod
> > characteristics, if I can get it tamed.
>
> Why not a 6EH7 there? That eliminates any question of matching AGC
> characteristics (with the IF stage tubes) and the 6EH7 is a
> superlative large signal performer, especially at higher bias (AGC)
> voltages. The gain is less than the 6EJ7 but that too is in the right
> direction.
>
> If the RF stage is fixed tuned (and must thus be broad) then much of
> the gain reduction can be done by loading the tuned circuits as in the
> HR-10. (Maybe a combination of loading and stagger-tuning to give a
> sharper drop-off outside the band you want to cover.) If it's tunable,
> then you want as high a Q as you can get (since it can't be high
> enough to make tuning awkward) and gain reduction can be accomplished
> with cathode bias on the tube and mismatching, perhaps by link
> coupling to the two coils. Of course that complicates the
> bandswitching for an all-band receiver.
>
> Another variable is the 1st mixer: Since you have excess gain to play
> with you could think about a lower gain mixer; maybe a 12AT7, maybe
> even a 12AU7.
>
> > Will get back to this receiver in a week or two. I think it will work a
> lot
> > better if I can, indeed, rethink the front end.
>
> It sounds like that's the only major issue.
>
> Walt Hutchens
>
> ______________________________________________________________
> HBR mailing list
> Home: http://mailman.qth.net/mailman/listinfo/hbr
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:HBR at mailman.qth.net
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
>