[Elecraft] K3. RX too sensitive for me - S-meter/RFgain/absolute mode...

[Ron D'Eau Claire]

Guy, you obviously didn't read my post.

I said the signal path was analog all the way to the second I.F. 

I agree that generating a stable, low phase noise local oscillator signal
digitally is great or as you might say "smart".  But that's not in the
signal path. Only the analog result - an r.f. signal - is used in the signal
path. 

What else is "smart" -- aided by digital logic -- is the development of that
analog bias voltage that is used to control the gain of the I.F. stages.
That is "smarter" than all-analog receivers, but it, too, is not in the
signal path.

And, yes, the pots and push-buttons go through logic to control the various
analog stages, but they are also not in the signal path. 

Behind all that glitzy and useful digital logic on the front panel is an
analog receiver -- all the way to the second i.f. 

Ron AC7AC

-----Original Message-----

On Wed, May 18, 2011 at 5:29 PM, Ron D'Eau Claire <ron at cobi.biz> wrote:
> Don't forget that the K3 *is* a largely analog radio. The signal path only
> goes digital at the second I.F.
>
> I believe the K3 (and K2) handle manual RF gain control just like any
analog
> radio does - they insert an adjustable d-c bias gain control voltage that
> replaces voltage originally created by the received signal.

Well, not so much analog as you think. ALL of the smart stuff is done
digital, including the frequency synthesizers.  NONE of the gain
devices are DIRECTLY controlled from a knob or a switch.  You can
easily verify this with some time in the schematics, which are out in
the open on the Elecraft web site.

The RF and AF gain pots all have a standard voltage on top and the
wipe in the pot is fed to an analog to digital converter (ADC).  The
ADC's now digital advice from those pots is passed to the CPU.
Firmware decides what voltage should be sent to the variable gain IF
stage and can completely ignore the RF and AF pot's numerical advice
if it wants to.  All the audio output devices are FIXED GAIN.  The
conversion from the number stream representing the audio is fixed.
The audio gain pot's numerical advice is interpreted by firmware and
the output of the speakers was a flat multiplication of the number
stream fed to the speaker's digital to analog converter chip (DAC).

There is no RF or AF pot voltage lead off the front panel boards. It's
buried in multiplexing done by the FPU and buried in the number soup
passed to the main CPU.

The PRE and ATT button states have to be interpreted by the CPU to set
an internal state for each of those.  The CPU then sends a state
"message" to a demultiplexer which puts high or low on an output pin
to appropriate lines which throw diode switches which either include
or exclude the preamp and attenuator altogether from the signal path.
This procedure is why such things can be changed from commands coming
in on the CAT connection.

There is a defensive hardware (analog) AGC which will keep the output
of the 15 kHz IF from overranging the last IF's and RX signal ADC with
very loud signals. This is the only conventional analog gain device,
and if PRE and ATT are being used properly, it will infrequently be
used to throttle the variable IF gain.  It has nothing to do with the
menu-driven firmware AGC processing.  Wayne's description of this is
that the main AGC simply assumes that any defensive AGC assertions on
the signal are coming in on the antenna.  And the result is as if
magically no signal coming in ever overranges anything.

A K3 has close to as little analog stuff as is possible, second only
to a Flex which goes digital one step closer to the antenna. But the
Elecraft method saved having a PC with expensive top end audio
devices.  Pretty sound trade-off IMHO.

Yes it does have some analog circuitry in it, but it sure ain't your
daddy's analog radio.

73, Guy.