Radio: it's not just a hobby, it's a way of life

Current Operating Frequency and Mode


Much improved propagation overnight as the Bz holds firm; Some areas report record precipitation static; Trans-Atlantic reports return to typical distributions; Good CW signal WA2XRM; TI3/W7RI providing reports from Costa Rica

– Posted in: 630 Meter Daily Reports, 630 Meters

Compared to what we have observed recently, this was a pretty good session although a number of areas are experiencing extreme cold with snow and ice.  While the extreme cold contributes to improved base currents on our vertical antennas, the ice and snow often create their own special problems.  In addition to mechanical issues from the precipitation, it also generates high noise levels.  Snow is particularly bad but rain can just as easily complicate listening.  Joe, NU6O / WI2XBQ, located in Northern California reported torrential rains which were creating the worst static that he had ever observed in that area.  Areas of the central and eastern US are beginning to feel the effects of a strong snow storm that is no doubt increasing the noise floor for many operators.


12-hour North American lightning summary


In spite of the elevated noise, propagation was really very good, at least from my perspective.  Judging by the level of activity and the number of stations that successfully provided reports, I would say that the noise didn’t spoil all of the fun overnight.

Part of this improved propagation seems to be the result of the Bz not pointing to the South for extended periods, thus protecting the magnetic field from the harmful solar wind that would pull its layers away like an onion and seriously increase absorption.  Not everyone enjoyed good propagation, however, as higher latitudes see the least benefit from this protection and variations in the Bz still result in some solar wind having an impact.  The early part of the evening session in North America experienced unsettled geomagnetic conditions and the solar wind velocities averaged very close to 700 km/s.  DST values were mixed but generally residing at negative levels.







Yesterday it was reported that Paul, W0RW / WA2XRM, would be QRV with a CW beacon on 479.9 kHz.  Paul’s beacon frequency, or very near it, is plagued with a nasty carrier and listening in the past has been challenging.  This night Paul’s signal was QSO quality and generally poking out from under the carrier very nicely.  His beacon was sending “VVV VVV VVV de WA2XRM WA2XRM WA2XRM COS”, where COS is the airport designator for Colorado Springs.  I have asked Paul to consider a cross-frequency QSO sometime.  I think we have worked in the past but his power increase has made his signal very easy to copy in the presence of the carrier.  Paul also posted the following session redux on the 600-meter research group email reflector:

” WA2XRM Reception Reports for 6 Jan. 2017.  The cold Wx  +10F and snow d’Q’ed  my tree slung top loaded vertical. I could not get better SWR than 1: 2.5.
Reports received so far:
1. AA7U, Steve, in the Desert of AZ,  coping my CW at 0037z and 0053z fading in and out.
2. KB5NJD, John, Dallas TX, Receiving at about 0130z, he has a near by annoying carrier but has filtered it out.  That is why I operate below 480 kHz.
3. W4OP, Dale, TN, could see  QRSS on ARGO, no copy by ear.
4. AD0VC, Tony, Denver, ground wave 70 miles away, good copy S4.
5. KU7Z, mark,  in UT got me on an 80 meter dipole, he did sound card capture and got my CW on his computer screen, his regular antenna coax is broken.
Others listening: W8TIC  MI,  but no report.”

Another interesting receiving opportunity resulted in reports being submitted by Scott, TI3/W7RI, located in Costa Rica and providing reports for WG2XXM and my station.  Scott writes:

“…on a whim I set my IC7200 to 630m and my WSJT-X program to WSPR, and on the first try, I decoded your signal beautifully, John.  Here is the decode:

0206  -19  -0.6    0.475793    0   WG2XIQ        EM12     37   2844

I am also decoding Kenneth’s signal on every transmit cycle.  Here is one of his decodes:

0212  -17  -0.8    0.475712    0   WG2XXM        EM15     37   3133

Besides your signals, I am seeing a third signal that is too weak to decode – it appears on even minutes, 100hz higher than your’s, Kenneth.  Both of your signals appear about 8dB above the noise floor on my waterfall.  Antenna is a G5RV at 50 feet, oriented to W7, with tuner bypassed.

I have to tell you I’m blown away by this.  I never expected it.  Down here in Costa Rica, I have an incredibly high powerline noise level on all bands (in excess of S9), and atmospheric noise that is truly brutal, as is typical for the tropics (though it is minimal this time of year).  So I didn’t expect to see anything but noise…”

Its really great to see DX activity like this and to receive reports like this one from Scott.  It goes to show what can be accomplished with what might otherwise be considered a simple system and it shows that noise won’t necessarily stand in the way of receiving signals below the broadcast band.  Thanks, Scott, for this detailed report!  Scott’s report details for this session can be viewed here.

Trans-Atlantic activity returned to more normal levels, as WG2XKA received reports from PA0RDT.  WD2XSH/17 was reported by G0LUJ and G8HUH, and N1BUG reported PA0A and EA5DOM.  Report details for these stations can be viewed here.

Doug, K4LY / WH2XZO, reports that he decoded ten WSPR stations and was decoded by 39 unique stations which are both improvements as his QRN has decreased, leaving band conditions in pretty good shape.  Doug adds that he  “…continues to be frustrated by not hearing Hawaii and some west coast stations as well. as well as they hear me, but Duke Energy has not repaired several of the worst line noise emitters.”

Rick, W7RNB / WI2XJQ,  reports that the previous session was significantly composed of reports with northeastern stations, this session was comprised of reports from stations in the southeastern US.  Perhaps this was in someway related to the weather system currently moving across the US.  Rick provided reports for nine WSPR stations and was reported by 33 unique stations.  Rick’s unique report details can be viewed here.

Mike, WA3TTS, reported that this session was an improvement and submitted these statistics:




Ernie,  KC4SIT (and soon to be WI2XQU), submitted the following comments and data for the session:

“Antenna, inverted L for 160 meters, is back in the air and working, always good. Conditions were lacking in Flat Rock NC last session with only 7 uniques. Perhaps the coming storms had some impact. Doug K4LY/WH2XZO, 32 km south of me had a 0 SNR where my previous spot of him was +15. Did not hear John KB5NJD/WG2XIQ until 01:52 UTC, several hours later than normal. No decode of western stations that I usually get. Here is my information. 73 Ernie KC4SIT”


KC4SIT session WSPR activity


Trans-Pacific report details for this session, excluding KL7 and KH6, are aggregated here.

Roger, VK4YB, reports, “Improved propagation tonight.  If the QRN would give me a break, I might be able to decode a new low power West coast station or a high power Mid West or East coast station. Just need a little extra lift.”  Roger received reports from JA1NQI, JA1PKG, JA3TVF, JA8SCD, JE1JDL, JH3XCU, JR1IZM, W7IUV, and he shared two-way reports with WI2XBQ.  Additionally, Roger provided reports for WG2XXM and WH2XGP.

Phil, VK3ELV, received reports from JA1NQI and JH3XCU.  Phil continues to receive decodes from JH3XCU and that data will be included in tomorrow’s report.

Ken, K5DNL / WG2XXM, reports that he was decoded by 74 unique stations, which was a record for his station, including reports from VK4YB, WH2XCR, and TI3/W7RI.  He received 81 decodes from WH2XCR,  best + 7 dB S/N, at a distance of 6007 km.    Ken reports three inches of snow on the ground and no problem with the antenna system match that might have otherwise rendered him off air due to high SWR.

Larry, W7IUV / WH2XGP,  received reports from sixty unique stations including JA1NQI, VK4YB, and VK2XGJ.  Larry provided reports for fifteen WSPR stations.  As W7IUV and using the western receive antenna, Larry provided reports to eleven WSPR stations including VK4YB.

Joe, NU6O / WI2XBQ, shared two-way reports with VK4YB.  In late comments, Joe noted that, “Great daytime prop to the north. Being spotted by CNF, MM, DBQ, XJQ, XSV, IUV. 690khz Vancouver BC booming in, hearing hetrodynes from JA BC stations in the truck this morning…Just logged in to my station by remote, 690 still @ +45db C/N”

I returned to WSPR in the late afternoon and experienced reasonable propagation through the evening.  There were certainly a large number of stations hearing well in spite of the various impedances that the session presented.  It was nice to receive reports from TI3/W7RI on WSPR as well as having good reception of WA2XRM on CW.  This morning the variometer was pretty well frozen up and I was short on time so WSPR continues to run during the daytime session as long as the match continues to hold.  My WSPR transmission report details can be viewed here and my WSPR reception report details can be viewed here.


WG2XIQ 24-hour WSPR activity


The band was abuzz through the session with very high and varied activity, although I failed to make a formal count.  Three stations were observed as new during this session including but not limited to KA2EKI, NC3Z, W3IH.  Welcome aboard!

Regional and continental WSPR breakdowns follow:


North American 24-hour WSPR activity



South American 24-hour WSPR activity



European 24-hour WSPR activity



Central / Asiatic Russian 24-hour WSPR activity



Japanese 24-hour WSPR activity



Australian 24-hour WSPR activity


Eden, ZF1EJ, provided reports for CF7MM, CG7CNF, VE3CIQ, WD2XSH/15, WD2XSH/17, WG2XIQ, WG2XKA, WG2XXM, WH2XCR, WH2XGP, WH2XNG, WH2XZO, WI2XBV, and WI2XJQ.  Report details for these stations can be viewed here.  Eden also operated as ZF1EJ/1, utilizing the new transmit antenna.  Report details for these stations can be viewed here.


ZF1EJ 24-hour WSPR activity



ZF1EJ/1 24-hour WSPR activity


Laurence, KL7L / WE2XPQ, reports that his receivers were off line overnight but his transmitter was alive, receiving reports from VK4YB and WH2XCR.  Report details for these stations can be viewed here.


WE2XPQ 24-hour WSPR activity


Merv, K9FD/KH6 / WH2XCR, reported VK3ELV and shared two-way reports with VK4YB.  He received reports from ZF1EJ, JA1NQI, JH3XCU and VK2XGJ.  WA3TTS also provided reports for Merv but most of that eastern transmitting stations were only reported by WH2XCR.  DX report details for this session can be viewed here.


WH2XCR 24-hour WSPR activity



“Today, let’s take a journey in 630m-land as if you owned imaginary receiving antennas that could separately receive 1-hop and 2-hop components of the same 630m signal. Suppose the simulation model (*Endnote 1) correctly generates and combines these propagation modes as illustrated in yesterday’s blog. Then nothing stands in our way from graphing those 1-hop and 2-hop components themselves!( **Endnote 2)

The upper-first illustration shows 1¼ simulated hours of 1-hop (blue) and 2-hop (purple) in dB. The model first forms electric field phasors of each hop mode, vectorially phases and combines them, and finally graphs their individual and combined signal levels in dB as shown in this first illustration.  I assumed m=20% amplitude modulation percentage of QSB by each sky reflection, a reference phase offset A=~100° between 475 KHz 1-hop and 2-hop, and slightly different rates of phase variation very close to 6mHz at the sky reflection points to get QSB curves as shown.  Remember that all these illustrated curves are QSB curves representing the envelopes of 630m RF.

Today’s second illustration brings in noise and establishes simulated 35% TxPct WSPR slots.  The combined signal level curves of each illustration resemble an exaggerated version of the 2-hop component rather than the 1-hop component.  I don’t yet know whether such exaggeration generally arises from the model, or incidentally is a feature of the specific model parameters used for the example.  Comparing both illustrations helps one to see how the mysterious slotted diversity of shapes in the second illustration neatly fits into the overall signal pattern of the first illustration.

You can see in both illustrations that the 1-hop component has just one main QSB frequency and less dB variation than the 2-hop component. Their combination into overall signal level magnifies the variability of the 2-hop component to produce occasional deep valleys of dB in overall signal strength.

If we could achieve a simulation model that convincingly reproduces an actual received signal by correctly generating the1-hop and 2-hop components beforehand, then by computerized means we might thus also have essentially provided that imaginary receiving antenna of 630m-land!”

*ENDNOTE 1: Form of the Model: The model calculates the real-value part of a vector sum of phasors that in general are each expressed as K e.  Parameters  S10, S20 set the 1-hop and 2-hop amplitudes respectively, and the three modulation indices m0, m1, m2 are all real numbers as well.  The model supposes that the 630m signal has an electric field strength phasor equal to the vector sum of single-hop and double-hop contributions to QSB.

         Sphasor = S10ejA (1+ m0 ejθo(t)) + S20 (1 + m1 ejθ1(t) )(1 + m2 ejθ2(t)) ejωt

Angular frequency ω is 2π times an individual 630m TX station’s frequency in Hertz, e.g. 475,700 Hz. Phase factor ejA recognizes or supposes that the 1-hop and 2-hop signals may be displaced by a constant phase. The constant phase is further supplemented by individual phase factors θ0(t) ,θ1(t) ,θ2(t) that account for the time varying nature of QSB.

      I write in place of S10ejA and S20 these expressions:

S10ejA /(1+cm0))  and  S20 /[(1+cm1)(1+cm2)] 

The model replaces simple amplitude constants with these amplitude-reducing denominators based on modulation indices m so that sharp dips (“sharps”) will generally lie at the lower amplitude level.  Unlike an amplitude modulator circuit, “amplitude modulation” by QSB probably does not increase the reflected signal strength compared to a perfectly horizontal ionospheric reflecting contour surface.  A noncritical parameter c ~3 to 5 (instead of unity) seems to work quite satisfactorily in the model, which would correspond to some additional reduction in reflectivity in the ionosphere by QSB.

Beyond this point, development of the model straightforwardly expands it in vector algebra, ignores the 630m reference frequency to focus on the QSB envelope itself, and takes the real part for column-by-column implementation in a spreadsheet such as Excel (Sh.3):

        SQSBreal = Re{(S10ejA + S20) + S10 m0 ej(θo(t)+A) + S20 (m1 ejθ1(t) + m2 ejθ2(t))

                                                                                                     +S20 m1 m2 ej(θ1(t) ej θ2(t))}


      SQSBreal = (S10cosA + S20) + S10 m0 cos(θ0(t)+A) + S20 m1 cosθ1(t) + S20 m2 cosθ2(t) )

                                                                                             +S20 m1 m2 cos(θ1(t)+θ2(t)) )

6 milliHertz average QSB frequency corresponds to 6.48 degrees of phase change per 3-second Echo mode sample spacing, indexed i.  I fitted the simulation satisfactorily by using these phase functions in the spreadsheet version of the model:

θ0(t) = 6.48°i;           θ1(t) =  6.48°i  x 19/17;          θ2(t) = 6.48°i  x 19/23.

The phasing strategy pictured a 1-hop mid-path sky reflection introducing 6mHz 1-hop QSB. By contrast, 2-hop propagation would have sky reflections geographically far-separated from each other on either side of the mid-path.  For physical reasons, I supposed the QSB frequencies of those two reflections would respectively be slightly greater and slightly less than the mid-path QSB frequency.  Prime numbers as above were used in the fractions to spread out a lot of different kinds of overall QSB behavior in simulation across a night of Echo mode samples.

Likewise for physical reasons, I set all modulation percentages equal and moderately small:

m0 = m1 = m2 = m = 20%.

Next, the above QSB electric field strength waveform SQSBreal is squared and converted to power in dB:        S(dB) = TxPwr(dBm) + 20 log10 SQSBreal      

Simulated Gaussian noise with zero mean and 1 dB standard deviation gives ~5dB noise spread, like what is observed on a good 630m night. Noise Level positions the noise relative to the signal.

         Noise N(dB) =NoiseLevel(dB)+NORM.INV(RAND(),0,1)

The noise is added to the signal at the level of power and then converted back to dB:

(S+N)dB = 10 log10(100.1S(dB) +100.1N(dB))

To visualize a simulated 630m WSPR signal sent at approximately 35% TxPct, I use nested modulo-prime functions operating every two minutes, i.e., every 40 samples of 3 sec. each. I call this “Picketed Noisy Signal”:

Picketed (S+N)dB =  If {[(└(i/40)┘)mod11 ]mod7}mod5 = 0 then (S+N)dB else N(dB).

Index i counts the samples, which are divided by 40 and rounded down to the next lower integer.  In Excel terms, I wrote:

IF(MOD(MOD(MOD(ROUNDDOWN(D2/40,0),11),7),5)=0, Y2,AB2)

At this point the calculations are completed. The Picketed Noisy Signal is graphed as if it were actual observed data by using Excel’s menu “Insert” and its first scatter diagram option.

**ENDNOTE 2: Simulation as if the antenna could separate 1-hop and 2-hop:

      SQSB1HOP =  S10  + S10 m0 cos(θ0(t)+A)                 (cosA factor omitted)

      SQSB2HOP =  S20 + S20 m1 cosθ1(t) + S20 m2 cosθ2(t)  +S20 m1 m2 cos(θ1(t)+θ2(t)) )

I convert each 1-hop and 2-hop mode to dB, provide picketing, and graph the curves on the same graph.  The cosA factor is omitted because the imaginary hop-specific receiving system of 630m-land would be insensitive to phase of each uncombined 1-hop or 2-hop signal component.   As a guesstimate, add (TxPwr(dBm) less 3dB) to the results for each of the two propagation modes, based on an assumed-equal power split between them.




Additions, corrections, clarifications, etc? Send me a message on the Contact page or directly to KB5NJD gmail dot (com).