Probably like you say... 1 crash per second of QRN. Slightly more than the
number of Orion crashes per day :*)
Using a constant carrier of 1uv mixed with the antenna input, when the crash
occured, the carrier was instantly and completely zapped, masked or blipped
out. It was as if the carrier had been turned off. Setting the AGC as I
mentioned seemed to be the only cure. I would estimate the crash levels to
be about S9 on the Orion's meter for what that's worth.
I remember having a Kenwood TS-570 and it did exactly the same thing to cw.
No setting of the agc did any good. At least with the Orion you have a
----- Original Message -----
From: "Sinisa Hristov" <email@example.com>
Sent: Monday, May 24, 2004 10:38 PM
Subject: Re: [TenTec] Orion AGC setting for QRN
> Steve Ellington wrote:
> > I ran the test and published the results. 2000 DB/S is exactly where it
> > worked best. As a matter for fact 2000 DB/S is the only place it worked.
> > Your statement indicates just the opposite. All I can say is "try it".
> > theory isn't always right or maybe QRN is different on this side of the
> I certainly trust your results and would like
> to understand what is actually going on, i.e.,
> I'd like to go a step beyond simply following recipes.
> Depending on QRN-to-signal ratio, it takes some time
> for AGC to restore RX gain after a QRN burst.
> For an arbitrary ratio of 40 dB (what's your figure?),
> it would take a mere 20 ms to fully restore the gain.
> A delay several times as long as that should be
> quite tolerable.
> What was your frequency of crashes, i.e. ~1 per second,
> a dozen per second, or a few per minute?
> Therefore, my point was that with the actual (vs. indicated)
> AGC decay rate of 2000 dB/s, and everything else unchanged
> (AGC attack time in particular), the system is probably too fast.
> On the other hand, one may pose the question of accuracy
> of the actual AGC rate. Orion already has some rather
> inaccurate selections (AGC threshold, CW rise and fall times,
> etc.), so one more is certainly possible :-(
> Sinisa YT1NT, VA3TTN
> TenTec mailing list
TenTec mailing list