"Michael A. Terrell" wrote in message
...
Tom Holden wrote:

"Michael A. Terrell" wrote in message
...
What do you consider "Fast" or "Slow"? The receivers we built had
seven different AGC time constants, up to 10 seconds.

.01 ms
.1 ms
1 ms
.01 second
.1 second
1 second
10 seconds

Michael, what receivers were these and why such incredibly fast AGC?
I assume these are release speeds; what were the corresponding attack
speeds?

Microdyne 700 series, and other lines like their 1100, 1400 and 2800
series. They were telemetry systems, and they were typically used in
diversity mode so you needed the fast AGC for some applications. They
had matching integration times for attack and release. I should know, I
tested hundreds of the boards on the bench while using a time interval
counter and calculating the reciprocals on the slower ranges.

Equal attack and release times over the entire range seems so
unconventional, at least for HF, MF, LF radios. I see the Microdyne 700 is
VHF/UHF and does not support AM. Fading at VHF/UHF line-of-sight is probably
very different than for ionospheric propagation as are the modes supported
and the fact that a human is not decoding the output.

From my readings, it seems that Fast Attack (less than 10 ms and closer to 1
ms often touted) was preferred for HF/MF/LF AM/SSB with selectable Release
speeds from a few tens of ms to over a second. After experimenting with AGC
modifications with that design objective, I'm inclined to think that the
Attack should be proportional to the Release, say, 10-20 times faster, not a
constant 10 ms. A really fast attack and really slow release combo resulted
in a single impulse of interference knocking gain down for a long time.
Equal attack and release speeds when too slow result in ear-damaging swells
of volume or unmanaged bursts of interference.

What do you think?

Any idea what the attack/release speeds are on your SW radios?

Tom