Propagation seemed OK overnight as corroborated by several operator reports for this session. I noticed more WSPR decodes than the previous session but S/N was considerably worse, limited to JT9 levels at my station for skywave paths. Storms continue to be active in pockets around the US but fortunately the local weather was less prolific than yesterday morning and I was able to complete a morning CW sked with few problems hearing.
Geomagnetic activity is similar to what has been seen over the previous few sessions where the GMF is quiet and the Bz was variable with elevated solar wind velocities above 400 km/s. Overall the trend has been toward calmer conditions.
John, WA3ETD / WG2XKA, makes a comeback with transcontinental reports through noisy conditions in Vermont:
Phil, VE3CIQ, also enjoyed transcontinental propagation as he tested a new receive antenna:
Neil, W0YSE/7 / WG2XSV, operated WSPR at a low transmit cycle overnight and still had nice reports for the effort in addition to receiving WH2XCR:
A number of stations have started their preparations and improvements for the next season. Steve, VE7SL, reports on his blog that he is working on a 630-meter transverter in hopes of completing a JT9 QSO with Roger, VK4YB. This Spring Steve observed significant openings on the path between British Columbia and Australia at levels that would easily support JT9 and in many cases CW also. If completed I believe that this would be the longest two-way QSO on record for 630-meters. It will be an epic achievement.
Al, K2BLA / WI2XBV, reports that he is working on a new 630-meter vertical and radial system that won’t have to be taken down each Spring for mowing. I’ve missed having Al QRV so it will be nice to have another active CW op that hears well.
Doug, K4LY / WH2XZO, has recently made antenna improvements that have paid off as he was decoded here through the noise at my station. As reported yesterday Doug has a number of station projects planned for the summer.
I hope to add the low noise receive vertical detailed on the blogs of VE7SL and WG2XKA. While I prefer to have some directivity in my receive antennas, it can be complicated when living in the center of the operator population and trying to give everyone a fair opportunity to be reported. I have tried a smaller, unbalanced version of this antenna and it did not seem much better than the E-Probe hanging in the Pecan tree but even that antenna has its place. I now have the proper toroid cores for balanced transformers and CAT cable to interface with the shack so this will be tested. Everyone that has used one of these receive verticals seems to be pleased with the performance.
Regional and continental WSPR breakdowns follow:
There were no reports from the trans-Atlantic or trans-African paths. UA0SNV was present during the session but no reports were observed.
Eden, ZF1EJ, continues to decode my station and WH2XZO. Rumor has it that Eden may have his station ready for first light shortly.
On the surface, Laurence, KL7L / WE2XPQ, seems to have had a very similar session to the previous. WH2XCR returned with two-way reports.
Merv, K9FD/KH6 / WH2XCR, returned for this session and like Laurence, appears to have had a very similar session to what he has had recently. The path to Australia appears to be open with reports from VK2XGJ and two-way reports with VK4YB.
Jim, W5EST, presents “PART 2: HOW SIGNAL AND NOISE VOLTAGES AND POWER COMBINE”:
“Today, let’s talk about how noise combines with other noise and how waveforms combine with each other generally. Noise is a big deal on 630/2200m and we try to obtain the most favorable signal-to-noise ratio SNR that we can.
For our purposes, and with apologies to statistical gurus,* let me start with what I mean by two waveforms being uncorrelated or correlated. I’m looking for insights that might promote noise reduction or cancellation.
If, for each instantaneous value in one of two waveforms, the other waveform’s voltage (or current) is equally likely to be positive or negative in value for every magnitude of the voltage (or current) you name, the two waveforms are uncorrelated. When either waveform is the same as the other one multiplied by an appropriate positive or negative scaling factor, the two waveforms are correlated. (In this blog post, “correlated” does NOT require use of a so-called correlator circuit.) You can cancel correlated waveforms with a phaser or canceller circuit, and can’t cancel uncorrelated waveforms.
Correlated waveforms add their rms values. 2+1=3, 3+2=5, etc. Subtract rms to do partial or complete cancellation: 2-1=1; 2-2=0.
By contrast, neither adding nor subtracting a first waveform to/from a second waveform that’s uncorrelated with the first waveform makes them any less uncorrelated. You just get some waveshape with additional rms either way. A sum of squares is involved because power P = Irms2R = Vrms2/R, and summing various power contributions means summing squares of currents or voltages. A square root “sqrt” returns you to rms of the combined uncorrelated currents.
Uncorrelated “2+1” = sqrt(22+(+1)2) = sqrt(5) = 2.36.
Uncorrelated “2-1” = sqrt(22+(-1)2) = sqrt(5) = 2.36, same thing.
For our purposes, sine waves of different frequencies, and non-overlapping spectra in general, are also uncorrelated.
A correlation coefficient r (or Greek letter rho ρ) numerically represents the degree to which two waveforms are correlated. Correlation coefficient r resembles the idea of phase as regards a sine wave. However, the correlation idea is not only pertinent to sine waves but also to other waveshapes besides sine waves. And noise certainly is not a sine wave.
Two waveforms V1 and V2 can be divided into correlated and uncorrelated parts relative to each other. Similarly, sine waves of identical frequency can be divided into 90° phased apart waveforms (uncorrelated, r=0) and remainder waveforms in-phase (correlated, phase 0° or 180°, r=+/-1.0). If a waveform V1 is a mixture of correlated and uncorrelated waveforms relative to waveform V2, then a value for a correlation coefficient r takes a value somewhere between 0 and plus or minus 1.
To get the power in a resistance R like a 50Ω receiver, think Irms2R. You get the power produced by the sum of two current waveforms I1 and I2 as follows:
1) Square the sum of the rms values of the correlated currents themselves.
2) Add the squares of the rms values of the uncorrelated currents to get another sum.
3) Take the total of sum (1) plus sum (2).
4A) Multiply the total from step (3) times resistance R.
P = R [ ( I1corr + I2corr)2 + I1uncorr2 + I2uncorr2 ]
4B) For voltage waveforms V1 and V2, it works the same way except divide the total by R.
P = (1/R) [ (V1corr+V2corr)2 + V1uncorr2 +V2uncorr2 ]
If the correlated waveform portions might have opposite sign, the correlated portions will cancel power out of each other when the waveforms are combined.
V1corr= -V2corr means V1corr + V2corr = 0.
Think of one waveform’s rms as negative in that case, if you like.
If you can provide two correlated versions of a significant portion of the noise power that’s in the receiver bandpass, you can reduce or cancel the correlated noise power portion by in-phase subtraction or 180° out-of-phase addition. More about noise cancellation in another blog post!
* https://en.wikipedia.org/wiki/Pearson_product-moment_correlation_coefficient (See illustrations, for instance. My definition is sufficient for our radio purposes, but some other wiki-illustrated instances also yield no-correlation r=0. I also assume the MF/LF signal and noise waveforms have zero DC level.)”
Additions, corrections, clarifications, etc? Send me a message on the Contact page or directly to KB5NJD gmail dot (com)!