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

Current Operating Frequency and Mode

OFF AIR for storms, probably for much of the week if the forecast holds

Better propagation reported by some operators while others report extreme storm noise as geomagnetic activity peak at G1 storm levels again; Trans-Pacific opening remain strong; W5EST presents: ”Solar Eclipse MF Simulation at W5EST: Why and What”

– Posted in: 630 Meter Daily Reports, 630 Meters

The details for August 19, 2016 can be viewed here.

IMPORTANT REMINDER: Neither 630-meters nor 2200-meters are open to amateurs in the US yet. That includes stations using fake or pirated call signs. Please continue to be patient and let the FCC finish their processes. UPDATED: Click here to view the proposed “considerate operators” frequency usage guide for 630-meters under Part-97 rules that was developed with the input of active band users.

The Atlantic and Gulf coast regions were a focal point for lightning producing storms during the evening and overnight.  Parts of the Midwest also experienced periods of active weather that contributed to elevated noise levels.  Alberta and Saskatchewan were impacted by a system that continues to produce lightning this morning.  Mexico and the Caribbean also experienced strong weather that contributed to elevated noise levels.

11-hour North American lightning summary


Geomagnetic conditions peaked again at G1 storm levels.  The Bz is pointing slightly to the South this morning and solar wind velocities peaked above 730 km/s while currently averaging at 670 km/s.  DST values have deteriorated again after showing signs of moderate improvement during the previous session.  Unsettled to storm levels are expected to continue for a while longer as a coronal hole continues to rotate across the path of the Earth.



Neil, W0YSE/7 / WG2XSV, provided reports for six WSPR stations including VE7BDQ, WG2XXM, WH2XCR, WH2XGP, WH2XXP, and WI2XJQ  but notes that he  missed WE2XPQ in Alaska again.  Neil added that he “…was surprised to be heard in Iowa this session, by W0JW. He spotted me 2 times at -28 each.  Out of 15 unique spotters, here are the ones greater than 1000 km distant from Vancouver WA”:


Doug, K4LY / WH2XZO, reported that “with less QRN, 29 stations decoded WH2XZO who decoded 7, both the best in a couple of months and included two stations over 3000 km, WH2XGP copied and VE6XH copying.”

Rick, W7RNB / WI2XJQ, provided reports for seven WSPR stations and he received reports from thirteen unique stations.


09:48 26XSH 0.475703 -29 0 CN98pi 0.1 WI2XJQ CN87ts 140 243
07:02 WH2XCR 0.475618 -26 0 BL11je 1 WI2XJQ CN87ts 4287 38
06:00 VE7BDQ 0.475736 -3 0 CN89la 0.5 WI2XJQ CN87ts 147 160
05:24 WG2XXM 0.475717 -27 0 EM15lj 5 WI2XJQ CN87ts 2499 311
03:06 WH2XXP 0.475663 -28 0 DM33 50 WI2XJQ CN87ts 1771 337
02:26 WG2XSV 0.475761 -26 0 CN85rq 2 WI2XJQ CN87ts 232 3
02:20 WH2XGP 0.475689 -13 0 DN07dg 5 WI2XJQ CN87ts 208 286


11:28 WI2XJQ 0.475611 +7 0 CN87ts 5 W7IUV DN07dg 208 105
10:14 WI2XJQ 0.475611 -20 0 CN87ts 5 VE7BPB CN89lg 174 344
10:08 WI2XJQ 0.475610 -17 0 CN87ts 5 WH2XCR BL11je 4287 239
10:08 WI2XJQ 0.475610 -17 0 CN87ts 5 WW6D CM88pl 1034 182
10:08 WI2XJQ 0.475610 -29 0 CN87ts 5 VE6JY DO33or 944 42
09:56 WI2XJQ 0.475611 0 0 CN87ts 5 VE7BDQ CN89la 147 341
09:40 WI2XJQ 0.475610 -8 0 CN87ts 5 WG2XSV CN85rq 232 183
09:10 WI2XJQ 0.475610 -16 0 CN87ts 5 WB6HYD CM87xi 1159 179
08:52 WI2XJQ 0.475610 -24 0 CN87ts 5 WE2XPQ BP51ip 2287 322
08:32 WI2XJQ 0.475611 +5 0 CN87ts 5 WH2XGP DN07dg 208 105
07:10 WI2XJQ 0.475610 -22 0 CN87ts 5 N7IW CN85mh 277 189
07:10 WI2XJQ 0.475611 -5 0 CN87ts 5 VE6XH DO24tc 899 35
06:50 WI2XJQ 0.475611 -28 0 CN87ts 5 W7WKR CN97uj 162 104

David, N1DAY / WI2XUF, reported that “It was a good night on 630M with the bulk of my activity occurring after 03:00 UTC.  21 stations heard my signal, and I spotted 7 stations.  I installed my upgraded main inductor coil yesterday and eliminated 8 turns over the old coil to bring the antenna to resonance.  At this point, the plan is to experiment with one variable every few days and observe reports.  Next up is modification of the top hat system.

Attached is a picture of the new coil – it is 12″ in diameter, 11″ tall (at the point of tapping), and is producing approximately 100 uH according to the calculators – about the same as the smaller coil with more turns.  The top and bottom forms are from an empty wire spool that Lowe’s gave me, and it just so happened that schedule 40 1/2″ PVC was a tight fit in the holes already cut into the form.”

New coil and feed point at WI2XUF


Ralph, W0RPK, is gathering and processing baseline data ahead of the eclipse:

“During 18Aug17, the last day for collecting 630m WSPR eclipse test data, 15-21z, we had 9 reporting stations and 208-reports.  I (allegedly) have the collection process under control and am ready to collect 630m WSPR eclipse study data for 5-days, 19-23Aug17, 15-21z.

Collected WSPR data for the control-group days, 19-20Aug17 and 22-23Aug17, will be compared with Eclipse day data, 21Aug17, looking for sky-wave propagation during the 15-21z window.

Hopefully we will have more 630m WSPR stations on-the-air.  Again on 18Aug17 there were no stations on-the-air east of Arizona.

During the eclipse I will also be listening for AM Broadcast station sky-wave propagation from 50kw stations. See http://www.skyandtelescope.com/2017-total-solar-eclipse/how-to-hear-the-solar-eclipse/.

Also during the eclipse I will be collecting output data from a small solar panel with obscuration peaking at 92% here in eastern NC.  Two discussions may be of interest: https://fivethirtyeight.com/features/the-solar-eclipse-vs-solar-electricity/ and https://ourfiniteworld.com/2017/07/22/researchers-have-been-underestimating-the-cost-of-wind-and-solar/.

WSPR reporting stations on 18Aug17
KF7NP   60-reports   WH2XXP 87km
N7IW    32-reports   WH2XGP 331km – WG2XSV 53km
WG2XSV  19-reports   WH2XGP 279km
WI2XJQ  32-reports   WG2XSV 232km – WH2XGP 208km
WH2XGP  12-reports   WG2XSV 279km
W7WKR   20-reports   WH2XGP 46km
VE7BDQ  29-reports   WG2XSV 373km – WH2XGP 315km
VE7KPB   3-reports   WH2XGP 382km
VE6JY    1-report    WH2XGP 868km”

Dave, N4DB, indicated a “Pretty good session last night despite the QRN which did quiet down substantially.” He reported that he decoded seven WSPR stations including “XXC XGP XXP XZO XFI XXM and XUF no Eden which is very unusual, likely he didn’t run last night”.

Trans-Pacific report details, excluding KL7 and KH6, can be viewed here.

Hideo, JH3XCU, submitted this link detailing DX -> JA decode totals and DX -> JA S/N peaks for the session, as reported on the Japanese language 472 kHz website.

Roger, VK4YB, reported “moderate QRN, good TP propagation. The winds arrived as forecast. No antenna repairs and 160m antenna is down, so I’m going backwards.” He received reports from E51WL, JH1INM, TNUKJPM, VE6XH, VE7BDQ, W7IUV, WH2XGP and WE2XPQ. He shared two-way reports with WH2XCR.

Ken, K5DNL / WG2XXM, reported that he decoded six WSPR stations and he was decoded by fifty unique stations including ZL2AFP, WH2XCR, ZF1EJ, and five Canadian stations.

Ward, K7PO / WH2XXP, received reports from 54 unique stations including E51WL, ZL2BPU, ZL1AFP, ZL4EI, VK4YB, VK2XGJ, VK3ALZ, and VK5AKK.

WH2XXP 24-hour WSPR activity (courtesy NI7J)


Larry, W7IUV / WH2XGP, provided reports for nine WSPR stations and he received reports from 34 unique stations including ZL2AFP and E51WL.  He shared two-way reports with VK4YB.   As W7IUV, Larry provided reports for nine WSPR stations, including VK4YB.

WH2XGP session WSPR activity (courtesy NI7J)


Regional and continental WSPR breakdowns follow:

North American 24-hour WSPR activity


South American 24-hour WSPR activity


European 24-hour WSPR activity


Chinese 24-hour WSPR activity


Japanese 24-hour WSPR activity


Oceania 24-hour WSPR activity


Eden, ZF1EJ, provided reports for four WSPR stations.  He noted that his absence from transmitting was due to error.

ZF1EJ 24-hour WSPR activity


Warwick, E51WL, provided for six WSPR stations by 1500z.  Those report details, excluding WH2XCR, can be viewed here.

E51WL 24-hour WSPR activity


Laurence, KL7L / WE2XPQ, provided reports for six WSPR stations including VK4YB. He shared two-way reports with WH2XCR.  DX report details can be viewed here.

WE2XPQ 24-hour WSPR activity


Merv, K9FD/KH6 / WH2XCR, provided reports for eleven WSPR stations including VK5ABN and VK5FQ. He shared two-way reports with VK4YB and WE2XPQ.  Merv received reports from 21 unique stations including E51WL,  EJTSWL, JA1NQI/2, VK2XGJ, VK2EIK, VK3ALZ, VK5AKK, VK7TW, ZL1BPU, ZL4EI and ZL2AFP.   DX report details can be viewed here.

WH2XCR 24-hour WSPR activity (not listed:  VK2EIK)



“On 630m, the work on antennas, equipment and operating skills comes first. Then comes the studious challenge of selecting potentially useful material from the volumes of information that operators upload to the WSPR database.  How do we make sense of  all that information and plumb the mysteries of 630m?

One way looks at the SNR sequences for patterns, peaks, and unexpectedly missing decodes. Supporting such effort is solar eclipse propagation simulation. Such simulation involves  math modeling of MF propagation effects of the eclipse. Hams and experimenters have lots of different specialities, and a simulation guy like me can share the work with the rest of the 630m community. Here’s why I think simulation is useful.

Simulation is a useful detective tool for insight into the mysteries of a 630m band that’s replete with mysteries.  Simulation means somebody sits down and step-by-step writes their understanding of usually several connected processes by which some aspect of 630m propagation works.

Simulation remembers and connects together the knowledge that was put into it. My imperfect memory certainly doesn’t do as well.  My thinking about something as complicated as 630m propagation can’t keep all the pieces of the puzzle in mind like simulation repeatedly can connect.

Simulation plays. The simulation executes the process pieces built up to constitute it. Out comes a whole set of dynamic information that can make graphs and visuals.  Simulation can add its own dimension to our sense of awe and wonder at the 630m workings of our Earth and Moon in the huge Solar System in which we live.

Simulation forces me to recognize when my understandings of 630m propagation are wrong or incomplete.  When actual 630m propagation delivers significantly different observations than the previously written simulation predicts, something’s wrong. Maybe storms got in the way, maybe antenna problems or equipment problems got in the way. But 630m people are relatively experienced and use the ON4KST reflector to say when that happens.  Most likely, when something’s wrong in the sense of departing from simulated predictions, the simulation is written wrong or my preconceived notions behind the simulation are wrong.  Then I’ve learned something!

What does the solar eclipse simulation have inside it, to serve the 630m community?   In the spreadsheet format I used, simulation  starts with a spreadsheet page to computes the geographic positions of RF-absorptive D-region crossings  based on the geographic positions of two particular 630m stations that you choose. (Endnote 1*) Since the D-region crossings are probably the culprits responsible for daytime 630m RF signal absorption, I want to know where they lie.

From the D-region crossing positions, a second spreadsheet takes over.  It’s loaded with tabular information downloaded last May specifying where the solar eclipse will be centered every few minutes of UTC time. The latitudes and longitudes change with each time row since the eclipse speeds from the NE Pacific Ocean, across the USA lower 48, and over the central Atlantic Ocean almost to the west African coast this Monday 8/21.

Next, the simulation figures partial eclipse percentage at the D-region crossings.  The percentage represents obscuration of the Sun’s disc area at those D-region crossing positions from each time to next time. (Endnote 2**).

From the varying levels of eclipse percentage, simulation estimates what’s happening at each D-region crossing. These estimates evolve and depend on time of day Zulu tZ translated to each D-region crossing longitude.  Simulation uses “244” as Aug. 21 day of year relative to winter Solstice in a 365.25 day year. That’s because the elevation of the rays of the Sun at solar noon at that geographic position depends on which day of the year it is. Sun elevation ϕ is calculated from the D-region crossing latitude L, and 23.4° Earth axis tilt with respect to the plane of its orbit around the Sun.

The simulation now proceeds to estimate RF absorption. which can be interpreted reversely as estimated SNR relative to nighttime.  (Endnote 3***).  No attempt is made to model whatever variations in E-region reflectivity may be present, nor to model phasing self-interference by multiple reflections from the E-region. The simulation effort focuses on D-region absorption in the two places where the RF signal presumably crosses it.

The calculations use obscuration and sine of sun elevation angle to involve a relative value of ionizing flux (watts / meter2) relative to nighttime. 630m RF absorption at a D-region crossing is assumed to increase with the Sun-induced ionization there, but with a twist. The simulation makes three different estimates of RF absorption based on the idea that absorption might respond solar ionizing flux like an RC circuit responds to like a low pass filter. These estimates involve alternative time constants  0 (none), 15, and 30 minutes.  Lastly, absorption dB at each of the two D-region crossings are summed together to provide each point for path graphing eclipse-effect dB on 630m RF signals.

Complicated?  You said it! Hopefully a bit like a solar eclipse is complicated! I’m glad to let the simulation do some of the hard work.  On Monday, station after station of diligent 630m folks will see what fruits of your own hard work come from the 630m mystery band.  TU & GL!

*ENDNOTE 1:   630m SOLAR ECLIPSE SIMULATOR AT W5EST mrsocion@aol.com
For further information and an example of results refer to June 2 kb5njd blog and its links.  http://njdtechnologies.net/060217/   . This solar eclipse simulation is available on request.  (MF sunrise simulation is a different project with different tentative parameter values and blogged in the Endnote at: http://njdtechnologies.net/081717/ )
The solar eclipse simulation  starts with a spreadsheet page that computes the geographic positions of RF-absorptive D-region crossings for single hop.  There, one starts by entering the latitude and longitude  (L1, G1),  (L2, G2) of each station at ends of a 630m radio path, in latitude (-90° to +90°) and east longitude (0°-360°).  To get the station positions, I magnify a WSPR map to identify 630m stations of interest and see what town in a state or  province they’re located. Then I web search on keywords for town, state, latitude and longitude to get the latitude and longitude of each path end.
The calculation assumes daytime E-region hE = 80km altitude (user-adjustable) and D-region halfway up ( hD/hE = 0.5) and then calculates path length fraction k  at which the 630m ray enters D-region (usually 0.25 to 0.35, but could reach 0.5 on near vertical (NVIS) paths.  To get path fraction k, it first calculates great circle g.c. 630m path distance D/RE , radian measure:
 D/RE =  arcsin{ sqrt[sin2L2 cos2L1 – 2 sinL2 cosL1 cosL2 sinL1  cos(G2 – G1) + cos2L2 sin2L1 
                                                                                                       + cos2L1 cos2L2 sin2(G2 – G1) ] }
Steps now compute a half-hop angle β with vertex at center of the Earth, radius  RE = 6371 km.
     β = D/(2RE).
Figure angle δ between RF launch ray and a straight line through ground between path ends:
     δ = arctan{[(1-cosβ)+ hE/RE] /sin β}  
Finally, obtain path fraction k from the above precursors:
     k = (RE/D) { β – δ + arccos[(cos(β – δ))/(1+hD/RE)] }    
Steps next get geographic position of each D-region crossing. First is calculated the tilt angle ζ (zeta) of the RF path g.c. plane relative to the equator:
    cos ζ = cosL1 cosL2 sin(G2 – G1)
Next to be calculated is longitude G0 where that RF path g.c. intersects the equator:
    G0 = arctan{ (cosL1 sin L2 sinG1 – cosL2 sin L1 sin G2)  /
                            cosL1 sin L2 cosG1 – cosL2 sin L1 cos G2) }
This leads to RF path-end longitudes G1’, G2  in primed RF path g.c. coordinates:
    G1’ = arctan(tan(G1 – G0)/ cos ζ ),
    G1’ = arctan(tan(G2 – G0)/ cos ζ ).
Given path fraction k, convert back to get first and second D-region crossing longitudes:
    GD1 = G0 + arctan[cos ζ tan(G1’+k(G2’– G1’) ]
    GD2 = G0 + arctan[cos ζ tan(G2’-k(G2’– G1’) ]
Analogously, calculate the first and second D-region crossing latitudes:
    LD1 = arcsin[sin ζ sin(G1’+k(G2’– G1’) ]
    LD2 = arcsin[sin ζ sin(G2’-k(G2’– G1’) ]
On a separate spreadsheet page for the eclipse simulation, one manually enter the positions (LD1, GD1), (LD2, GD2) for the RF-absorptive D-region crossings.  An eclipse timing map tells us when maximum partial eclipse may occur at various geographic positions sideways from the track of totality.  https://eclipse2017.nasa.gov/downloadables  (Scroll 10% and click big blue download button.)
The Aug. 21 eclipse simulation was preloaded with eclipse coordinates for all times along eclipse track in May, 2017 by download from this now-revised site: https://eclipse.gsfc.nasa.gov/SEpath/SEpath2001/SE2017Aug21Tpath.html (I scrolled 20% for Table, col. 3 Central Line (i.e., (Le , Ge) : eclipse totality track latitudes & longitudes in columns vs. times in rows).) It also tells the eclipse magnitude Mt at totality for each position, which varies from 1.016 to 1.030 to 1.015 along the whole eclipse totality track. Moon diameter is
 2 Rmoon = 3474km.      
Eclipse magnitude M is the partially eclipsed Sun diameter fraction. Totality is translated by the spreadsheet to partial eclipse magnitude M at the D-region crossing location by this long formula that includes a square root of a sum of three squares:  M =
=Mt[1 – (RE/(2Rmoon) sqrt{[cosL1 sin(360°(-1+ tZ/24) +G1) -cosLe sin(360°(-1 + tZ/24) + Ge) ]2
                                       + [(sinL1 – sinLe) cos[arctan[tan(23.43703°) cos(360° 244/365.25) ]]]2
+ [(cosL1 cos(360°(-1+ tZ/24) +G1) – cosLe cos(360°(-1 + tZ/24) + Ge))
                                                         x  sin[-arctan[tan(23.43703°) cos(360° 244/365.25) ]  ]2 } ]
As a function of eclipsed Sun diameter fraction M, the solar eclipse obscuration Obsc fraction of sun area in eclipse is: Obsc = (2/π)[ arcos(1-M) – (1-M)sqrt(2M-M2) ].  Obscuration Obsc (M) is separately calculated for each D-region RF crossing for each 2 minutes during the Aug. 21 eclipse.  As a function of time at a given geographic location, Obsc(t) is approximately a triangular shaped function going from 0 to a maximum partial eclipse fraction and then back to 0.
On some paths, partial eclipse obscuration occurs before or after the times the pre-loaded total eclipse data begins or ends.  Fortunately, partial eclipse obscuration is a time-symmetrical function that can be manually completed by copying and reverse-sorting some data from the opposite time-end of the triangular obscuration graph data.
To estimate varying solar ionizing flux on D-region, the sine of sun elevation angle ϕ is calculated for each D-region crossing depending on time of day at given latitude, longitude, and Aug. 21 day of year. Its implicit assumptions of spherical earth and circular earth orbit are ok for our radio purposes.  The first D-region crossing is used for example.
sin ϕ = sin(L1) sin{-arctan[tan(23.43703° )cos(360° 244/365.25) ]}
           – cos(L1) cos{arctan[tan(23.43703°) cos(360° 244/365.25) ]} cos[360°(-1 + tZ/24) + G1]
From the varying levels of eclipse percentage, simulation of absorption involves a concept of solar ionizing flux at each D-region crossing by using obscuration. In spreadsheet row after row, Obsc(t) and sin ϕ(t) help calculate a relative value of ionizing watts/m2  relative to nighttime.  The absorption formula also assumes that the D-region density varies with the ionizing flux.  Jointly, solar ionizing flux and D-region density deliver an absorption formula labeled V0(t) as if it were to drive an RC circuit. An assumed -30 dB adjustable value of noon vs. night relative RF absorption means each D-region crossing has up to -15 dB absorption (adjustable) within which its absorption formula works.  After derivation and fitting, the formula I put into the simulation was:
         V0(t) = 1+1015dB/10 sqrt[(1- Obsc(t))*sin ϕ(t) /sin(32.6°)]
Absorption at a D-region crossing is assumed proportional to the ionization there except with a time constant of 0, 15, or 30 minutes which is subsequently introduced in respective spreadsheet columns. If no time constant, resulting absorption V(t)  = V0(t) .  Absorption itself, and not its dB, is what is low-pass “RC filtered” with 15 or 30 minute time constant.
The low-pass “RC filter” with 15 or 30 minute time constant τ is applied according to this next formula. With 2 minutes between spreadsheet rows, “2/ τ” is written in the formula.
         V(t+1) = V(t) + (1/(1+1/τ))*((V0(t+1)+V0(t))/2 –V(t))*(2/τ)  ; initial value V(1)=V0(1).
Absorption dB at each zero crossing is computed by the usual 10 log10() calculation from time-evolving absorption V(t) resulting from the “RC filter”  The formula is computed for each D-region crossing respectively, and absorption dB is only then calculated as
Absorption dB = 10 log10(Absorption) for each crossing. Finally, the absorption dB at the two D-region crossings for each moment of time are summed for path graphing the eclipse effect in dB on 630m RF signals.

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