As I understand the graphs and operation: AGC Threshold is the point at which AGC action starts. The higher the AGC THR number, the higher point at which AGC starts. The highest AGC THR (8) starts AGC action at S3. That is a pretty low level but any setting less than THR 8 starts the AGC action with even less signal.
AGC SLP is how aggressive the AGC action is as signals rise. A highest AGC SLP number means the AGC flattens everyone. AT SLP = 15 all signals above S3 are brought to roughly the same level. Heard one at a time, two signals - one at S3 and one at S9 will sound the same level. But, since AGC works on every signal in the passband, if there are two signals - one S9 and the other S3 appearing at the same time, the S3 signal will disappear because the AGC is knocking down the S9 signal to an S3 level and thereby also reducing the S3 signal to the noise. Using a less aggressive (lower) slope number will help but the dynamic range between and S3 and S9 signal is compressed. The core problem is the AGC starts at such a low level. Background noise alone on 80 - 40 is almost always more than S3. The suggestion is an AGC THR setting to allow higher than an S3 level before AGC action starts. That makes sense but does it start to degrade the high-end dynamic range of the radio? Are we in danger of overloading the intermediate amplifiers/mixers generating IMD and other unwanted issues? Seems like there is a balancing act here. Another way to accomplish this is to use the Attenuator and/or reduce the RF gain to reset our starting point so that the S5 signal at the antenna becomes an S3 signal at the AGC level and our AGC action doesn't really begin until S5. I would like the ability to switch in even more attenuation for high noise conditions. Can that be done through the software? Still I would like to try a higher AGC THR setting and see how it plays. The beauty of this radio is its ability to adjust to different conditions - one size does not fit all. Unfortunately, its also complicated. It has taken me three years to figure this little bit out. Comments? Do I understand it right? Buck k4ia K3 #101 In a message dated 12/9/2010 10:11:19 A.M. Eastern Standard Time, [email protected] writes: Thanks Don. I just realized that I can do some quick experiments to get my answer using an XG2's s3/s9 capability and noting audio level increase with AGC Off and then with AGC On and reducing RF Gain until same audio increase is accomplished. An aside to KE7X and W4TV: I too wish we could try experimental firmware that significantly raised the max AGC THR value. And to K6LL: Yes, I too fear loss of weaker signals 73, Barry N1EU On Thu, Dec 9, 2010 at 3:00 PM, Don Wilhelm <[email protected]> wrote: > Barry, > > Without doing any measurements it may be difficult to say precisely, and I > would think there would be a bit of difference from one K3 to another. > > However, there is a quick and easy way to accomplish that - note that the > S-meter increases as the RF gain is reduced. So if you want to increase the > AGC Threshold by (say) 3 S-units, disconnect the antenna, and reduce the RF > Gain until you move the no-signal S-meter up by 3 S-units. That should be > close to your goal. > > 73, > Don W3FPR > > On 12/9/2010 9:07 AM, Barry N1EU wrote: > >> >> Cady, Fred wrote: >> >>> I made some measurements and have posted the results for your perusal >>> and information at www.ke7x.com. >>> >> Very interested findings - Fred has a graph of AGC action with AGC THR=8 >> labeled figure 3 on http://www.ke7x.com/home/the-k3-agc-story : >> >> So I ask this group: is it possible to raise the AGC Threshold from s3 to >> s6 >> or s9 by simply reducing the RF Gain control? Can anybody estimate what >> settings of the RF Gain control would accomplish that? >> >> Barry N1EU >> >> ______________________________________________________________ ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[email protected] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html
