[Elecraft] AGC/S-meter

Fri May 9 14:40:00 2003

I started figuring out how the meter ought to respond and how it is responding,
and the following might help depict what I'm thinking/observing

-----

LED 1    2    3    4    5    6    7    8    9   10    Off-the-scale
   S1             S5*       S9       +20       +40

  -121dBm        -97dBm    -73dBm    -53dBm    -33dBm          (Ideal)

                 -85dBm    -73dBm                       -53dBm (Reality)

*(I believe that S5 probably ought to sit between LED 3 and LED 4, so I won't
seriously consider S5's ideal value.)

-----

At best, I can hope to get "S1" and "+40"  set for ideal - if "S9" works out,
GREAT! (but I doubt it will happen)

My first step will be to analyze what signal strengths give a reading of S1 and
+40 (as well as each LED in between).

Then I'm going to insert a -121dBm and adjust S-Lo for a reading of S1 (and
adjust my AGC from 3.80 V if needed).

Next I'll try to insert a -33dBm and adjust S-Hi for a reading of +40.

Then I'll re-analyze what signal strengths generate various LED bar readings.
That might help determine what amount of gain exists between each reading and
if the meter reading is 'linear' (e.g. 20dB between each LED, or if it takes
more dB as the meter goes higher?).

- Daniel/AA0NI

--- "Guy Olinger, K2AV" <[email protected]> wrote:
> The only scalable S meter I have seen in my entire life was my Heathkit
> SB300, and that with a caveat. Using a precision attenuator, each major scale
> line on the meter was just about 10 db, which was not what the scale said. 
> 
> Using just barely starting to move meter as S0. S3 was +10db, S6 was +20db
> and S9 was +30. 30 over 9 (full scale) was +60. 
> 
> Once I knew this, it was handy as a moderate accuracy comparison tool.
> 
> And I couldn't tell you if that was due to the Heathkit design or blind luck.
> 
> Moral of the story. S-meters are RELATIVE. 
> 
> Maybe we can get Wayne/Eric to come up with an add-on to the K2 which is a
> true scalable RF voltmeter, another choice in the frequency, date/time,
> battery voltage cycle. Or something to allow us to linearize the S meter the
> same way the VFO rate is linearized or such. With real 6 db s-unit ticks and
> 10 db ticks over S9, and individual band offsets to maintain accuracy over
> the spectrum.
> 
> Starts to sound complicated doesn't it...
> 
> 73, Guy
> 
> > 
> > From: Daniel Reynolds <[email protected]>
> > Date: 2003/05/09 Fri PM 01:24:08 EDT
> > To: [email protected]
> > Subject: RE: [Elecraft] AGC/S-meter
> > 
> > I found that out at lunch when I put the K2 on the bench at work. Running
> 50uV
> > into the K2 on 7040 yielded S9 (after I readjusted my S-meter last night...
> > more about that adjustment process below).
> > 
> > BUT - when I dropped from 50uV (-73dBm) to -53dBm (+20dB gain), the meter
> went
> > from S9 to Full Scale... and when I backed it off to -85dbm (approx 2
> S-units)
> > it actually dropped to S-5...
> > 
> > So the lesson is that the meter is very relative (and not very precise)!
> > 
> > -----
> > 
> > I adjusted my meter last night using the following procedure...
> > 
> > - AGC Voltage set at 3.80.
> > - Set S Lo with max RF gain
> > - Set S Hi with RF gain at 9 o'clock (instead of min RF gain).
> > 
> > It resulted in what I felt was significantly improved meter action compared
> to
> > what my meter was doing before (it was very stingy).
> > 
> > -----
> > 
> > Is it accurate? Is the K2 designed to be exactly 12-dB between each LED
> between
> > S-5 and S-9 and 10-dB between each LED from S-9 to +40 dB? Who knows...
> > 
> > I just know that I 'feel' better seeing the meter travel more than S-1 or
> S-2
> > on signals that seem pretty clear and 'strong' to my ears. I also know that
> a
> > strong signal (50uV) will now give me a strong reading of S-9.
> > 
> > BTW - 50uV on 7040 was S9, and 50uV on 21060 was S7 - so the gain varies
> from
> > band to band.
> > 
> > - Daniel/AA0NI
> > _______________________________________________
> > Elecraft mailing list: [email protected]
> > You must be a list member to post to the list. 
> > Postings must be plain text (no HTML or attachments). 
> > See: http://mailman.qth.net/mailman/listinfo/elecraft
> > Elecraft Web Page: http://www.elecraft.com
> > 
> 
> _______________________________________________
> Elecraft mailing list: [email protected]
> You must be a list member to post to the list. 
> Postings must be plain text (no HTML or attachments). 
> See: http://mailman.qth.net/mailman/listinfo/elecraft
> Elecraft Web Page: http://www.elecraft.com