For those of you following the saga of my AGC obsession, here's the latest.
I installed Dave's mod with only one change. I left C3 connected to pin 5 of U2 rather than moving it to the RF Gain line as Dave's diagram shows. I have emailed Dave about this, but haven't heard back yet. The mod works very well. I adjusted the AGC line to have maximum gain from the MC1350 with the RF Gain control fully clockwise and minimum gain from the MC1350 with the RF gain control fully counter clockwise. The RF gain control now controls the entire AGC range of the MC1350. This measured 68 dB of gain range on my K2. In addition the difference between AGC on and off is now a little less than 0.6 dB. Before the mod it was a little more than 0.8 dB. An improvement of about 0.25 dB. I tend to be one of those operators that uses the RF gain control to adjust listening levels. With this mod the AF gain can be set at maximum, and with the RF gain at minimum the receiver is virtually silent. Nice work on the audio section Wayne and Eric! I never expected it to be this quiet. For those implementing Dave's AGC mod, I'll go into more detail of my adjustment procedure as it differs slightly from what Dave suggests. First off I determined the AGC line voltage that gives me maximum gain from the MC1350. This is measured at pin 1 of U2. The primary tool used was an audio analyzer. My PC is equipped with an M-Audio Audiophile 2496 sound card and uses Baudline as audio analysis software. I suppose this is similar to Spectrogram on Windows computers, but I have never used Spectrogram. The K2 was set to 28.002 MHz, attenuator on, AGC fast, CW mode, SSB OP1 filter selected, and RF gain at maximum. A DL1 dummy load was connected to the antenna jack. The audio signal to the PC was tapped off the speaker output with an 8 ohm speaker connected. The AF gain on the K2 was adjusted to produce a noise floor around 40 dB down as displayed in the Baudline spectrum analyzer. As one adjusts the RM2 pot, one can see on the spectrum analyzer, the approximate point where the gain of the MC1350 hits maximum. The AGC line voltage at this point was used as a starting point. As the MC 1350 nears maximum gain, the change in gain slows, so it is difficult to see the exact point of maximum gain. Setting Baudline to measure the total power of the received noise floor from the K2, and averaging this reading over 5000 FFT transformations yielded a reading stable to within 0.05 dB. The AGC line voltage was then set to a few increments of 100 mV above and below the starting point voltage and the audio power level was measured at each increment. This narrowed down to the 100 mV range where maximum gain was reached. I then continued to narrow down the voltage by halving the voltage difference until I found the point of maximum gain. On my K2 this was 3.325 volts on the AGC line. Interestingly this coresponds to a measured AGC Threshold voltage of 3.43 volts. As the accuracy of my voltmeter is not calibrated, for reference my voltmeter reads the 8v rail as 8.24 volts and the 5v rail as 4.98 volts. Next, I needed to determine the minimum gain AGC voltage. The noise floor of the K2 is not sufficient to find this level. An Elecraft XG1 set to output a 50 uV signal on 7.040 was used. The K2 was set to 7.040, the preamp may be necessary to get a good reading. The audio signal was viewed on the spectrum analyzer and RM2 adjusted to find the point of minimum gain. This is nowhere near as critical to determine and a voltage of 5.10 volts was measured on the AGC line. This results in a total swing of 1.78 volts on the AGC line from maximum to minimum gain of the MC1350. Following Dave's procedure for adjusting RM1 and RM2 resulted in the RF gain control changing the AGC line through the full range of 3.325 to 5.10 volts. Dave's procedure for adjusting RM3 and the s-meter worked well. I set the S HI setting to show S9 on the meter with a 50 uV signal from the XG1. So far I am very happy with this modification. I can no longer hear any difference in the noise floor with the AGC switched between on and off. I also have an RF gain control with a far greater range. The gradual onset of AGC action that I like so much in the K2 seems to be preserved. I wonder if Jack Smith, K8ZOA, of Clifton Laboratories could be enticed into performing this mod and posting the new AGC graph. If others try this modification please post your results. 73, Darrell VA7TO K2#5093 -- Darrell Bellerive Amateur Radio Stations VA7TO and VE7CLA Grand Forks, British Columbia, Canada _______________________________________________ Elecraft mailing list Post to: Elecraft@mailman.qth.net You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com
