Optimum RF gain settings for WSPR

1 post / 0 new
G8JNJ
G8JNJ's picture
Offline
Last seen: 3 days 11 hours ago
Joined: 2008/05/19 - 12:32
Optimum RF gain settings for WSPR

Hi All,

I tried a few more experiments today, with another radio running WSPR on 160m

The idea is to find the best RF gain setting that will still give good, full range reporting of WSPR signals but by reducing the amount of AGC action occurring within the receiver, it will hopefully improve the accuracy of S/N reports.

I wanted to find a simple method of setting RF gain, which is repeatable and which others can easily apply to their station.

The first set of experiments on 160m indicated that the range of noise produced by my receiver and antenna was about 40dB. By turning the RF gain to minimum to give a 0dB reference, and then by gradually increasing the RF gain, making a note of the noise figure and resetting the soundcard RX gain back to zero each time, it was possible to determine the overall noise level with maximum RF gain relative to the 0dB level with Minimum RF gain.

Once I have measured the overall range. I set the RF gain back to minimum and reset the sound card RX gain to give me 0dB noise. I then increased the RF gain to give me 6dB of noise and once again reset the soundcard RX gain to 0dB noise. By increasing the RF gain to give increases of 6dB in noise level each time I was able to capture a few WSPR stations at each setting and compare the overall results.

With the two radios I have tried so far (Icom IC-7000 and a Yaesu FT-897D) it would seem that the optimum setting is with the RF gain set so that the noise level is approx 10dB above the minimum value of noise obtained with the RF gain control set to minimum. In both cases this corresponded to an S meter level of about S9.

So far I have only tested the two radios on the one band. One gave a 40dB range of noise the other a 30dB range. In both cases I found that I seem to get the best results with the internal pre-amp on, as this allowed me to reduce the RF gain control more than with the pre-amp off, which in turn helped to further reduce the AGC action.

Obviously this will only work where the majority of the noise is being introduced external to the receiver. On 10m for example I found the noise range to only be about 10dB, in this case it may not be possible to improve things any further by reducing the RF gain.

So, I urge folks to experiment with their RF gain settings to see if they can improve the S/N reports of the stations they are receiving (especially on the LF bands) and also examine if it is improving the consistency of reports over a successive series of spots.