[TenTec] Re: Al Gavenas' position on Front-End atenuation

[George, W5YR]

Thank you, thank you, thank you, Allan!

I have been corresponding with Al on this plus doing my homework plus
conducting some fairly interesting experiments using an advanced audio
analysis program. The bottom line for me is that with four radios on the
desk  the recommended procedure does nothing that I can perceive or measure
toward improving signal-to-noise ratio. And I would be greatly alarmed if it
did.

Following is the response I made to Al's original posting:

"I have an Icom IC-765, a 756 PRO, a 756 PRO2 and a K2 on the desk. I tried
your recommended approach and it does nothing beneficial on any of the
radios.

"Back in my youth (1946) when my National HRO-5TA1 was the hot stuff, your
approach
worked because there were no AGC systems that could work with CW signals.
Besides, with *two* high-gain r-f amps ahead of the mixer and marginal  AGC
(it was called AVC back then!)
at best, if you didn't back off on the RF gain - which really did control
the gain of the r-f amp stages - you could
overload them and get all manner of cross-mod, overloading, etc.

"I really think that this business of running the audio wide open and riding
the "RF" control is a holdover from the old days when it was applicable to
the radios of an era past. But, with today's radios, I really have to wonder
how it could do much good since the operator is primarily acting to defeat
the very carefully designed and implemented multi-loop, distributed AGC
control system.

"Our experiences seem to have been different, since I find with all four
rigs
mentioned above that letting the designer's AGC system do its job cannot
really be improved upon by my fumble-fingering a couple of controls.
Especially since receivers made in the past decade or so rarely have
anything in the r-f stages to control with the "RF Gain" control! "

Al has indicated that he will track down the QST reference for me.

The experiments I conducted just this afternoon are as follows:

With a noise generator driving the PRO2 to an S-meter reading of S8 on
40-meter CW, I injected a continuous carrier (RTTY) signal from the
companion PRO at the same S-meter level of S8.  With both signal and noise
present, the S-meter reading was S9+15 dB. The noise level within the 800 Hz
passband was -55 dB on the spectrum in the absence of the signal, while the
signal alone peaked at -33 dB.With signal present, AGC action reduced the
noise floor to -72 dB. The S/N measured by the program with an 800 Hz
passband in the PRO2 was 28 dB. The peak signal to noise level spectral
difference was 72-33 = 39 dB.

I then arbitrarily reduced the "RF Gain" (which controls the IF gain via AGC
system bias and the bias on a PIN diode *after* the first mixer - nothing in
the "r-f" section of the radio) such that I measured a noise level of -75 dB
from the noise generator, compared to -55 dB with max RF Gain setting. That
occurred with the gain control set to about 9o'clock. With the same signal
injected again, the peak signal was -37 dB for a difference of 38 dB between
peak signal and noise level. The S/N as measured by the software remained at
28 dB. All these values fall well within experimental tolerance.

So, Allan, I greatly appreciate your confirming the position that I had
taken with Al and in other exchanges on other reflectors. I fear that this
"reduce the r-f gain and full-blast on the audio" is about to become another
fetish similar to the 1:1 SWR craze. There are doubtlessly instances with
certain receiver designs under certain operating conditions where the
operator can do a "better" job with manual "r-f gain" control than can the
receiver circuitry, assuming that the operator even has control over "r-f
gain" with the knob so labeled. But, as you point out, this is 2003 and our
current radios just cannot be compared to or operated in the same manner as
our 1946 models, no matter how hard we try!  <:}

73/72, George
Amateur Radio W5YR -  the Yellow Rose of Texas
Fairview, TX 30 mi NE of Dallas in Collin county EM13QE
"In the 57th year and it just keeps getting better!"
<mailto:w5yr@att.net>







----- Original Message -----
From: "Allan Henry Kaplan" <w1ael@mindspring.com>
To: <tentec@mailman.qth.net>
Sent: Wednesday, April 09, 2003 5:17 PM
Subject: [TenTec] Re: Al Gavenas' position on Front-End atenuation


> The article to which Mr. Gavenas refers was, alas, one of the most
> misinformed and misleading pieces to ever appear in QST!  Sometimes you
> can decrease the effect of atmospheric noise as a distraction to a
> wanted signal by turning down the "RF Gain" or inserting an RF
> atenuator.  The catch is:  That strategy works only if the desired
> signal is substantially louder than the noise.  If the signal is weak --
> not a whole lot greater than the external noise -- RF attenuation (or
> gain control) will degrade the net signal-to-noise ratio.  If one runs
> the AF gain wide open and controls speaker-blasting with the RF gain, he
> emphasizes the receiver's internal hum and noise with respect to the
> weak signal.  In many cases, the weaker signals simply will not be
> heard!
>
> Running AF gain wide open and riding the RF gain control is just BAD
> ADVICE unless you are in a QSO with pretty loud signals -- KWs across
> town on 75 meters, perhaps.   If you are after DX or QRP that is the
> WRONG way to go.
> 73, Allan, W1AEL.
> _______________________________________________
> Tentec mailing list
> Tentec@mailman.qth.net
> http://mailman.qth.net/mailman/listinfo/tentec