Jan-Martin Noeding, LA8AK wrote:
On Mon, 23 Feb 2004 10:28:19 GMT, Doug Smith W9WI
wrote:

Tom Holden wrote:
In researching desirable AGC characteristics that might
be applied to the RS DX-394 over a year ago, I came
across the terms 'delayed' and 'hang'. Thought they
were interchangeable but on reading the ARRL 2004
Handbook, it seems that 'delayed' means that the attack
speed on the RF stages is slower than on the IF stages
or is relatively delayed. According to the HB, "This
prevents a premature increase in the receiver noise
figure".

I always thought "delayed AGC" means that there's no
gain reduction unless the strength of the incoming
signal reaches a certain threshold. It's not a delay in
time, but in amplitude.

Yes, this is correct for the proffesionals, a certain
amplitude level must be reached before the AGC threshold
is hit, while amateurs started to talk about time delay,
rise and fall times when AGC was optimized for SSB
reception. But even receivers designed before WWII had
some degree of mode-dependent time delay optimization,
fast for AM and somewhat slower for CW
I experienced the importance of proper time constant when
I tried to improve the Collins 51-S, see
http://home.online.no/~la8ak/b35.htm

I seized on the word "delayed" and linked it to the inclusion of a longer
time constant in the AGC to the RF stages of the DX-394 than the time
constant in the AGC to the IF stages. The Handbook example actually says "As
an option, the AGC to the RF amplifier is held off, or 'delayed', by the
0.6V forward drop of the diode so that the RF gain does not start to
decrease until larger signals appear. This prevents a premature increase in
the receiver noise figure. Also, a time constant of one or two seconds after
this diode helps keep the RF gain steady for the short term."

I think 'delay' is a misnomer if what we have in effect is a higher
threshold of signal strength for activation of RF AGC than for IF AGC.
'Two-step' AGC might be a better description. 'Delay' seems a more
appropriate term for the way AGC is implemented in the DX-394. RF stage
attack speed is slower (extra RC time constant probably on order of 100ms)
than that for the IF stage and is clearly 'delayed' in reaching steady
state, no matter what the signal level change is, as long as the AGC is
activated by the higher, later level. Release speed of the RF AGC is
similarly slowed or 'delayed' relative to IF AGC, certainly not by the 1 or
2 seconds in the HB example.

Having removed the 'delay' capacitor, I have not noticed any deleterious
effect. What should I be looking for? Would there be some advantage in
revising the RF AGC to achieve a 2-step AGC?

73, Tom

Tom Holden wrote:
[snip]
I concluded the following to be good targets for AGC
behaviour after surveying a number of radios -
corroboration or otherwise appreciated.

Attack: 1-13ms
Release:
- fast: 25ms
- medium: ~300ms
- slow: 1.8-3 seconds

I thought the fast release to be too fatiguing for human
listening to SSB speech and ICW code but desirable for
machine decoded data formats to minimise loss of data.
Also, with audio derived AGC, the distortion on heavy
bass modulation of all AM modes would be excessive.

In applying mods to the DX-394 by others and some
designed by myself, stretching the release time towards
the 'slow' target has the side effect of lengthening the
attack time to potentially a few hundred milliseconds. My
version is the fastest so far with an attack of about 100
ms on a release of 2 seconds. I'm wondering if there is
much to be gained by struggling to bring that down to the
target of 1-13 ms.

Comments on my assumptions, logic, conclusions and
questions most welcome!

Still looking for feedback on these targets.

Had one private reply that raised an interesting point about noise blanker
operation and 'delayed' RF AGC:
"You might want to test out a system of this sort in the face of a
significant amount of broadband impulse noise - e.g car-ignition noise
or something like that.

If the attack is made too fast, then impulse noise at a rate of, say,
60-100 Hz would tend to force the RF AGC into its low-gain state, and
this could allow weaker RF signals to be buried in the noise of the
first-stage mixer.

Using a longer attack time constant will render the RF AGC a lot less
vulnerable to the effects of impulse noise. You can then use a noise
blanker, located prior to the first IF AGC, to trim out this noise.
As long as the first-stage mixer and the RF amp stage aren't forced
into excessive intermodulation by the content of the impulse noise,
I'd expect that this approach would give you the best set of
behavioral tradeoffs."

73, Tom

Tom Holden wrote:
[snip]
I concluded the following to be good targets for AGC
behaviour after surveying a number of radios -
corroboration or otherwise appreciated.

Attack: 1-13ms
Release:
- fast: 25ms
- medium: ~300ms
- slow: 1.8-3 seconds

I thought the fast release to be too fatiguing for human
listening to SSB speech and ICW code but desirable for
machine decoded data formats to minimise loss of data.
Also, with audio derived AGC, the distortion on heavy
bass modulation of all AM modes would be excessive.

In applying mods to the DX-394 by others and some
designed by myself, stretching the release time towards
the 'slow' target has the side effect of lengthening the
attack time to potentially a few hundred milliseconds. My
version is the fastest so far with an attack of about 100
ms on a release of 2 seconds. I'm wondering if there is
much to be gained by struggling to bring that down to the
target of 1-13 ms.

Comments on my assumptions, logic, conclusions and
questions most welcome!

Still looking for feedback on these targets.

Had one private reply that raised an interesting point about noise blanker
operation and 'delayed' RF AGC:
"You might want to test out a system of this sort in the face of a
significant amount of broadband impulse noise - e.g car-ignition noise
or something like that.

If the attack is made too fast, then impulse noise at a rate of, say,
60-100 Hz would tend to force the RF AGC into its low-gain state, and
this could allow weaker RF signals to be buried in the noise of the
first-stage mixer.

Using a longer attack time constant will render the RF AGC a lot less
vulnerable to the effects of impulse noise. You can then use a noise
blanker, located prior to the first IF AGC, to trim out this noise.
As long as the first-stage mixer and the RF amp stage aren't forced
into excessive intermodulation by the content of the impulse noise,
I'd expect that this approach would give you the best set of
behavioral tradeoffs."

73, Tom

In article , "Tom Holden"
writes:

In researching desirable AGC characteristics that might be applied to the RS
DX-394 over a year ago, I came across the terms 'delayed' and 'hang'.
Thought they were interchangeable but on reading the ARRL 2004 Handbook, it
seems that 'delayed' means that the attack speed on the RF stages is slower
than on the IF stages or is relatively delayed. According to the HB, "This
prevents a premature increase in the receiver noise figure". On the DX-394,
I found and removed an electrolytic capacitor in the AGC line of the RF
stages that has the effect of slowing the attack relative to that of the IF.
Can't say I hear any difference with it out. I would have thought that we
would want the AGC attack to be fast on all stages in order to avoid
momentary overload.

According to terms used a half century ago in tube-type receivers,
"delayed AVC" (they used 'AVC' back then instead of 'AGC') just
meant that AGC action is delayed from taking effect until the signal
level rises above wide-open gain. Wasn't a time delay, but closer
to "hold off" voltage of some sort on the AGC control voltage action.

"Attack" and "decay" meant the time of response of the AGC, in
the same manner as audio-only control circuitry for broadcast
audio modulation input. "Attack" being the onset of a signal and
subsequent audio gain reduction time..."decay" being the release
time of any audio gain reduction control signal, that 'decaying' back
to original high-gain amplification.

I suspect that the original "delayed AVC" came about from two
reasons: First was that 'communications receivers' of olden times
usually tuned a rather wide band, usually 1:3 in frequency range
and the non-AGC gain varied by almost the same amount as a
result of tuned circuit impedance changes (tubes would amplify
almost entirely by only voltage changes when running Class A).
A hold-off on the AGC action handled the difference in overall gain
at the extremes of tuning.

Second, and I suspect the stronger reason, was that customers
could hear the front end noise in the speakers (with AGC held
totally off and gain wide open) and get the impression of a "really
hot receiver!" :-)

It's easy to get any kind of "delayed AGC" or hold-off on AGC
control with a DC bias on the detector.

As to the attack and decay times you specified, seems good to
me. That can be set to suit the individual listener.

Len Anderson
retired (from regular hours) electronic engineer person

In article , "Tom Holden"
writes:

In researching desirable AGC characteristics that might be applied to the RS
DX-394 over a year ago, I came across the terms 'delayed' and 'hang'.
Thought they were interchangeable but on reading the ARRL 2004 Handbook, it
seems that 'delayed' means that the attack speed on the RF stages is slower
than on the IF stages or is relatively delayed. According to the HB, "This
prevents a premature increase in the receiver noise figure". On the DX-394,
I found and removed an electrolytic capacitor in the AGC line of the RF
stages that has the effect of slowing the attack relative to that of the IF.
Can't say I hear any difference with it out. I would have thought that we
would want the AGC attack to be fast on all stages in order to avoid
momentary overload.

According to terms used a half century ago in tube-type receivers,
"delayed AVC" (they used 'AVC' back then instead of 'AGC') just
meant that AGC action is delayed from taking effect until the signal
level rises above wide-open gain. Wasn't a time delay, but closer
to "hold off" voltage of some sort on the AGC control voltage action.

"Attack" and "decay" meant the time of response of the AGC, in
the same manner as audio-only control circuitry for broadcast
audio modulation input. "Attack" being the onset of a signal and
subsequent audio gain reduction time..."decay" being the release
time of any audio gain reduction control signal, that 'decaying' back
to original high-gain amplification.

I suspect that the original "delayed AVC" came about from two
reasons: First was that 'communications receivers' of olden times
usually tuned a rather wide band, usually 1:3 in frequency range
and the non-AGC gain varied by almost the same amount as a
result of tuned circuit impedance changes (tubes would amplify
almost entirely by only voltage changes when running Class A).
A hold-off on the AGC action handled the difference in overall gain
at the extremes of tuning.

Second, and I suspect the stronger reason, was that customers
could hear the front end noise in the speakers (with AGC held
totally off and gain wide open) and get the impression of a "really
hot receiver!" :-)

It's easy to get any kind of "delayed AGC" or hold-off on AGC
control with a DC bias on the detector.

As to the attack and decay times you specified, seems good to
me. That can be set to suit the individual listener.

Len Anderson
retired (from regular hours) electronic engineer person

On Mon, 23 Feb 2004 21:33:39 -0500, "Tom Holden"
wrote:


I seized on the word "delayed" and linked it to the inclusion of a longer
time constant in the AGC to the RF stages of the DX-394 than the time
constant in the AGC to the IF stages. The Handbook example actually says "As
an option, the AGC to the RF amplifier is held off, or 'delayed', by the
0.6V forward drop of the diode so that the RF gain does not start to
decrease until larger signals appear. This prevents a premature increase in
the receiver noise figure. Also, a time constant of one or two seconds after
this diode helps keep the RF gain steady for the short term."

Mentioning a certain voltage level doesn't really make so much sense
when you don't know what the rest of detector stages looks like. Drake
2-B has 5Vpp from 6BA6, and R-4C somewhat less, but not too important.
This is only a reference which may not apply to any other receiver
because it is another AGC amplifier which also amplifies IF to the AM
detector for 2-B, and product detector fed via a voltage divider. For
R-4C everything is totally different, and the 4-5Vpp level seems more
chosen for economical reasons. Another English LF communication
receiver I checked had 80-100V pp IF output. The level to choose
depends on good AGC characteristic, and usually the IF level should be
certain level above the background noise, possibly 10-20dB for good
operation. Some receivers have not particularly good AGC, Atlas 210X
is one, and my Yaesu FT-902 is another, even worse is FT101B because
carrier oscillator leaks into the IF and AGC threshold must be set
10dB above the level you would want it to to start, simply because it
can't operate properly below it.
It was a surprise to learn that the Lorenz 6P203 receiver operated
nicely on SSB when BFO level was increased, in spite that it has only
6AV6 detectors, but IF is split, and it has two different 6BA6's with
different diode detectors, one for AM/CW/SSB and one for AGC.
http://home.online.no/~la8ak/b71.htm
Siemens Rainbow receiver also had some improper connection between
product detector and BFO, and could detect SSB well when this was
corrected http://home.online.no/~la8ak/b72.htm
Heathkit SB300/301 have bad AGC, but may be easily improved a lot by
using 1N4148 AGC detector diodes (voltage doubler), simply because the
original diodes have too much capacitance and do not rectify properly.
I did some experiments using Drake 2-C type and R-4B type AGC
detectors in my Kenwood TS-500, later in 2-B, and could set the
detector output variation from AGC threshold to as little as 1dB, but
it seem no real point since the subjective sound seem best with at
least 6dB variation, but wasn't too important if it was kept as
original 10dB for 2-B, it is only important when you have a lot of
advanced measuring equipment in the shack, but not at all for the
radio operator.

73, LA8AK


----
Jan-Martin, LA8AK, N-4623 Kristiansand
http://home.online.no/~la8ak/

On Mon, 23 Feb 2004 21:33:39 -0500, "Tom Holden"
wrote:


I seized on the word "delayed" and linked it to the inclusion of a longer
time constant in the AGC to the RF stages of the DX-394 than the time
constant in the AGC to the IF stages. The Handbook example actually says "As
an option, the AGC to the RF amplifier is held off, or 'delayed', by the
0.6V forward drop of the diode so that the RF gain does not start to
decrease until larger signals appear. This prevents a premature increase in
the receiver noise figure. Also, a time constant of one or two seconds after
this diode helps keep the RF gain steady for the short term."

Mentioning a certain voltage level doesn't really make so much sense
when you don't know what the rest of detector stages looks like. Drake
2-B has 5Vpp from 6BA6, and R-4C somewhat less, but not too important.
This is only a reference which may not apply to any other receiver
because it is another AGC amplifier which also amplifies IF to the AM
detector for 2-B, and product detector fed via a voltage divider. For
R-4C everything is totally different, and the 4-5Vpp level seems more
chosen for economical reasons. Another English LF communication
receiver I checked had 80-100V pp IF output. The level to choose
depends on good AGC characteristic, and usually the IF level should be
certain level above the background noise, possibly 10-20dB for good
operation. Some receivers have not particularly good AGC, Atlas 210X
is one, and my Yaesu FT-902 is another, even worse is FT101B because
carrier oscillator leaks into the IF and AGC threshold must be set
10dB above the level you would want it to to start, simply because it
can't operate properly below it.
It was a surprise to learn that the Lorenz 6P203 receiver operated
nicely on SSB when BFO level was increased, in spite that it has only
6AV6 detectors, but IF is split, and it has two different 6BA6's with
different diode detectors, one for AM/CW/SSB and one for AGC.
http://home.online.no/~la8ak/b71.htm
Siemens Rainbow receiver also had some improper connection between
product detector and BFO, and could detect SSB well when this was
corrected http://home.online.no/~la8ak/b72.htm
Heathkit SB300/301 have bad AGC, but may be easily improved a lot by
using 1N4148 AGC detector diodes (voltage doubler), simply because the
original diodes have too much capacitance and do not rectify properly.
I did some experiments using Drake 2-C type and R-4B type AGC
detectors in my Kenwood TS-500, later in 2-B, and could set the
detector output variation from AGC threshold to as little as 1dB, but
it seem no real point since the subjective sound seem best with at
least 6dB variation, but wasn't too important if it was kept as
original 10dB for 2-B, it is only important when you have a lot of
advanced measuring equipment in the shack, but not at all for the
radio operator.

73, LA8AK


----
Jan-Martin, LA8AK, N-4623 Kristiansand
http://home.online.no/~la8ak/

Jan-Martin Noeding, LA8AK wrote:
Mentioning a certain voltage level doesn't really make so
much sense when you don't know what the rest of detector
stages looks like. [snip]

Thanks for the interesting examples, Jan-Martin. The quote was from the 2004
ARRL HB page 17.23 in reference to a schematic/block diagram of a "typical
superhet receiver with AGC applied to multiple stages of RF and IF". The RS
DX-394 bears some resemblance to this with a resistor in place of the "Delay
Diode". It beats me how the AGC in this HB example actually controls the RF
gain - the diode appears to block control. A second one in parallel in the
opposite direction seems to me to be needed.

Tom

Jan-Martin Noeding, LA8AK wrote:
Mentioning a certain voltage level doesn't really make so
much sense when you don't know what the rest of detector
stages looks like. [snip]

Thanks for the interesting examples, Jan-Martin. The quote was from the 2004
ARRL HB page 17.23 in reference to a schematic/block diagram of a "typical
superhet receiver with AGC applied to multiple stages of RF and IF". The RS
DX-394 bears some resemblance to this with a resistor in place of the "Delay
Diode". It beats me how the AGC in this HB example actually controls the RF
gain - the diode appears to block control. A second one in parallel in the
opposite direction seems to me to be needed.

Tom

"Jan-Martin Noeding, LA8AK" wrote in message
...
On Mon, 23 Feb 2004 21:33:39 -0500, "Tom Holden"
wrote:


I seized on the word "delayed" and linked it to the inclusion of a longer
time constant in the AGC to the RF stages of the DX-394 than the time
constant in the AGC to the IF stages. The Handbook example actually says
"As
an option, the AGC to the RF amplifier is held off, or 'delayed', by the
0.6V forward drop of the diode so that the RF gain does not start to
decrease until larger signals appear. This prevents a premature increase
in
the receiver noise figure. Also, a time constant of one or two seconds
after
this diode helps keep the RF gain steady for the short term."

Mentioning a certain voltage level doesn't really make so much sense
when you don't know what the rest of detector stages looks like.
73, LA8AK


Jan-Martin,

I think the issue is that it is NOT a time delay, but a signal level
delay. The AGC does not begin to reduce receiver gain until the signal
reaches some defined LEVEL.

--
Steve N, K,9;d, c. i My email has no u's.