[Elecraft] Re: K1 AGC time constant [and how to radically improve K1 AGC]

[wayne burdick]

Keith wrote:

> Hmmm, well, I may be confused but I *think* you just told us how to
> convert the K1 from AF derived AGC to RF derived.

That's correct, Keith, assuming you think 150 kHz is "RF"  :)


> I realize making the AGC respond faster to a large signal would take
> that sort of mod, but isn't it just a cap change to speed up the decay
> after the large signal is gone?

No, and this is at the heart of the tradeoffs with AF-derived AGC. No 
matter how small you make the detector's capacitor, it still takes time 
for it to charge, and it must do so at the rate imposed by the 
bandwidth of the crystal filter. With a narrow-band filter, the 
resulting slew rate is long enough to result in an unpleasant few 
milliseconds of high-amplitude "pop" getting through the AF amp. The 
second problem is that a small detector capacitor won't filter out AF 
ripple from the signal applied to the detector, and you'll end up with 
AF modulation riding on the AGC signal. This creates objectionable AF 
IMD.

So-called "Hang AGC" and a very high-impedance detector can be used to 
minimize this effect, but you can't completely eliminate the "pop" this 
way, at least in my experience. An AGC signal derived from 150 kHz will 
be far superior, as anyone who has a K2 will attest to.


> Of course with no RF gain, I'm not sure I really want to speed up the
> AGC recovery.  In my TenTec rigs that were audio-derived, I'd speed up
> the AGC recovery then I'd control the leading edge overdrive by backing
> down on the RF gain manually.  Can't do that on the K1.

In practice, RF gain is rarely backed down by K2 users -- the AGC 
system has a very pleasant sound and very wide dynamic range. This 
could easily be achieved with the K1, too. But you could incorporate a 
manual RF gain control if you wanted to. Finding a place on the K1's 
front panel is another story.

Wayne


---

http://www.elecraft.com