16 February 2014

Power Whispering and Propagation

Normally, I am a shy guy who doesn't boost about my roaming around. But sometimes someone shows you that things which you are doing are more interesting than it seems to be at the first glance.  An so it was today: Bert (PA1B) wrote me, that he made a new analysis and if I was interested, I could post it on my blog. So here it is:

With permission of Bert (PA1B)
So what's so interesting in these figures?
Just to make it clear, I haven't sent my contacts to Bert. They are public and you could get them by applying a filter on the Wsprnet.org site or go to the download page. And if you want to do some propagation analysis you get so much data, that your computer will be busy for a while.

But back to the chart from Bert. You see some reports I have got. You already know the distance, call sign and the power I reported in my transmission. All these information are part of report. One important part of this data set is the SNR (Signal-Noise-Ratio). It compares the level of my signal at the antenna of M0NKA (see example below) to the level of the background noise.

When I make a QSO with my 5 Watt (37 dB) my signal should be at least 4 dB at the other station, so that the operator with a good receiver could hear (decode) my signal. (Look for normal SNRs on the Reverse Beacon Network.) That is a loss of 33 dBm. And this attenuation is caused by antennas and propagation. But remember: When you are working with normal modes, the SNR should be greater than 0 dB otherwise the operator on the other side will only listen to noise.

When whispering things are different. the coded signal could be understood (decoded) down to -33 dB. That is far below the hearing level and a good and valid report could be -29 dB. That is sufficient.

         Timestamp      Call    MHz    SNRDriftGridPwrReporterRGridkmaz
2014-02-16 14:34  DL1SDZ  14.097097  -14  -2  JN48mm  0.1  M0NKA  IO92am  895  304 
2014-02-16 14:24  DL1SDZ  14.097118  -15  -1  JN48mm  0.1  M0NKA  IO92am  895  304 
2014-02-16 14:14  DL1SDZ  14.097119  -13  -2  JN48mm  0.1  M0NKA  IO92am  895  304 
2014-02-16 14:04  DL1SDZ  14.097087  -12  -2  JN48mm  0.1  M0NKA  IO92am  895  304 
2014-02-16 13:24  DL1SDZ  14.097082  -16  -3  JN48mm  0.1  M0NKA  IO92am  895  304 
2014-02-16 13:14  DL1SDZ  14.097039  -18  -3  JN48mm  0.1  M0NKA  IO92am  895  304 
2014-02-16 13:04  DL1SDZ  14.097138  -13  -1  JN48mm  0.1  M0NKA  IO92am  895  304 
2014-02-16 12:44  DL1SDZ  14.097147  -17  -3  JN48mm  0.1  M0NKA  IO92am  895  304 
2014-02-16 12:34  DL1SDZ  14.097118  -13  -2  JN48mm  0.1  M0NKA  IO92am  895  304 
2014-02-16 12:24  DL1SDZ  14.097078  -22  -3  JN48mm  0.1  M0NKA  IO92am  895  304 
2014-02-16 12:14  DL1SDZ  14.097125  -20  -2  JN48mm  0.1  M0NKA  IO92am  895  304 
When you look at the SNR from my log they differ between -13 dB and -22 dB and my power was 100 mW or 20 dB. When I would have been interested to establish a connection with the least possible power I could have reduced my power by -10 dBm. I was sending with too much power 10 mW would have been enough to use the actual propagation. (Look at the data and you will find out when propagation was down.)
So this is what Bert is presenting in his charts. Let me cite:
I use the calculated lowest possible power in milliwatt of each WSPR spot, when
  • the spots are made with different power
  • the spots are all made over the same distance
The lower the calculated lowest possible power, the better the propagation.
To understand the mathematics please have a look at Berts Calulation .

All in all: I am a real QRO-GUY with my 100 mW. I still have to learn, that QRPp and Milliwatting with WSPR can be done with 1-10 mW, (if propagation is good).

So I hope that there are no mistakes in my explanation, and if so, please tell me and I will correct it.

But now let's turn to some music with better SNR.

Stay Tuned!