An Introductory Tutorial to W6ELProp

Running Your Own Propagation Predictions

Carl Luetzelschwab, K9LA April 2004

Are you interested in rolling your own propagation predictions? If you answered yes,
UHDGRQ7KHUH¶VDYHU\XVHU-friendly program out tKHUHWKDWDOORZV\RXWRGRWKLV,W¶V
IURP6KHO:(/DQGLW¶VFDOOHG:(/3URS:(/3URSLVWKH:LQGRZVYHUVLRQRIKLV
popular (but now discontinued) DOS-EDVHG0,1,3523VHULHV$QGLW¶VDIUHHGRZQORDG
WRR6ROHW¶VGRZQORDGLWDQGVHWXSWKHRSWLRQV7KHQZH¶OOUXQDSUHGLFWLRQDQGUHYLHZ
what the outputs mean.

Downloading W6ELProp is easy. Go to www.qsl.net/w6elprop. Scroll down a bit and

FOLFNRQ'RZQORDG:(/3URSYRUZKDWHYHUYHUVLRQLW¶VDWZKHQ\RXGRZQORDGLW
<RX¶OOEHGRZQORDGLQJDQDSSURximate 480K file called W6ELPropInst270.exe. Execute
LWDQGIROORZWKHGLUHFWLRQV<RX¶OOHQGXSZLWKWKH:(/3URSLFRQRQ\RXUGHVNWRS

Start W6ELProp and then click on the Options menu. In the Default Terminal folder,
enter your call and your latitude and longitude (from www.qrz.com, for example).

Next go to the Frequencies and Constants folder. For the frequencies, enter the
IUHTXHQFLHV\RX¶GOLNHSUHGLFWLRQVRQ,XVXDOO\XVH
and 28.3 (you can even add the new 60m frequency, too). Note that W6ELProp only
allows frequencies down to 3.0MHz. It does not do predictions on 160m. For 160m, I
usually note the times when the entire path is in darkness.

For Additive Signal Level Constants, this is where you input your station specifics in dB
referenced to 100w and a dipole. For example, if you have an older 1000w amp that does
80-PEXWH[FOXGHVWKH:$5&EDQGV\RX¶GDGGG%>ORJZZ@WRDOO
but the WARC frequencies. If you have a tri-bander on 20m, 15m, anGPWKHQ\RX¶G
DGGDQDGGLWLRQDOG%RQWKRVHIUHTXHQFLHVWKDW¶VURXJKO\WKHJDLQRIDVPDOOWUL-bander
over a dipole). The following figure shows the frequencies and additive signal level
constants for the above scenario.
In the Prediction Parameters folder, go with Minimum Radiation Angle = 1.0 degree (use
higher values if your QTH is blocked at low elevation angles), Noise Bandwidth =
3000Hz for SSB (use 500Hz for CW), Signal Level Suppression Threshold = 0dB, Man-
Made Noise Environment as dictated by your QTH (mine is Rural), and put a check in
the Suppress Zero-Availability Predictions box.

In the User Preferences folder, go with Sunspot Number for the Primary Solar Index,
Signals Levels for the Primary Signal Display, UTC for Time Display, 3 minutes for map
Auto Upgrade Interval, and Low for Frequency Map Resolution.

1RZFOLFNRQWKH6DYH0\6HWWLQJVDQG([LWEXWWRQ'RQ¶WIRUJHWWRGRWKLV<RXVKRXOG
now be back at the main screen.

Click on the Predictions menu. Then select On-6FUHHQ+LW³HQWHU´DQGWKHGHIDXOWFDOO

latitude, and longitude that you entered earlier should show up on the left side of the
window as Terminal A. For this example, my default data will be used.

/HW¶VUXQDSUHGLFWLRQWR$UJHQWLQD,I\RXGRQ¶WNQRZWKHSUHIL[click on Select from

Atlas and select Search by Name. Scroll down until you see Argentina. Highlight it, and
then click OK. The prefix, the latitude, the longitude, and the name should show up as
7HUPLQDO%)RUD'DWHOHW¶VJRZLWK'HFHPEHUZKLch is typed in using the
format MM/DD/YY.

For Solar Index, earlier we selected Sunspot Number as the Primary Solar Index. The
smoothed sunspot number should be entered here. Since the smoothed sunspot number is
a 13-month running average centered on the desired month, the official smoothed sunspot
number is 7 months behind real time. So how do we find the predicted smoothed sunspot
number for December 2003 (or for any future month)? One method is to go to
sec.noaa.gov/ftpdir/weekly/Predict.txt and use the value in the left-most column (which
is the PREDICTED smoothed sunspot number) for December 2003. From that table,
ZH¶OOHQWHUIRUWKH6RODU,QGH[1RWHWKDWWKHGDWDRQWKDWZHEVLWHDOVRKDVVPRRWKHG
solar flux. So if we had chosen Solar Flux as our Primary Solar Index, we could have
entered a smoothed solar flux.

Why should we use the smoothed sunspot number (or smoothed solar flux) and not the
daily solar flux? To develop the model of the ionosphere for our propagation predictions,
solar data was compared to ionosphere data. The best correlation was between the
smoothed sunspot number (or smoothed solar flux) and monthly median ionospheric
SDUDPHWHUV7KXVWKHPRGHOZDVGHVLJQHGWRWDNHDVPRRWKHGVRODULQGH[7KH³PRQWKO\
PHGLDQ´DVSHFWPHDQVWKDWRXUSUHGLFWLRQVDUHVWDWLVWLFDOLQQDWXUH:H¶OOJHWLQWRWKDWD
little bit later.

For the K Index, enter -1, which indicates the predictions will be for an average K index.

Now click on OK. After a couple seconds, you should see data specific to the chosen path
- the latitude and longitude of both terminals, sunrise and sunset for both terminals, the
bearings, and the short and long path lengths. Click on Show Predictions and you should
have a screen full of data like the following.

One of the most obvious features of this data is the difference between the lower
frequencies (80m, 40m, and 30m) and the higher frequencies (17m, 15m, 12, and 10m).
7KH\¶UHHVVHQWLDOO\GHJUHHVRXWRISKDVH7KLVLVEHFDXVHDEVRUSWLRQGULYHVWKHORZ
EDQGVVRWKH\¶UHEHVW only during the night, whereas MUF (maximum usable frequency)
GULYHVWKHKLJKEDQGVVRWKH\¶UHEHVWGXULQJWKHGD\

The first column in the figure is the time in UTC. The second column is the monthly
median MUF. Remember when I said our predictions are statistical in nature? What this
means is that the actual MUF is predicted to be at least as high as the monthly median
MUF on half the days of the month. For example, at 1700 UTC, the actual MUF on half
the days of the month will be at least 29.7MHz. On the other half of the days of the
PRQWKWKHDFWXDO08)ZLOOEHORZHUWKDQ0+]7KHELJSUREOHPLVZHGRQ¶WNQRZ
ZKLFKDUHWKH³JRRG´GD\VDQGZKLFKDUHWKH³EDG´GD\V

Each frequency column has two entries - a number and a letter. The number is the
predicted monthly median signal strength in dB above 0.5uv. Assuming S9 is 50uv and
an S-unit is 5dB (the generally accepted standard), the dB values in W6ELProp translate
to S-units as follows:

Signal level in dB S-units

50 10dB over S9
45 5dB over S9
40 S9
35 S8
30 S7
25 S6
20 S5
15 S4
10 S3
5 S2
0 S1

Being a median value, on any given day of the month the signal strength could be
somewhat above or below this predicted value.

The letters A, B, C, and D refer to the probability (in terms of the number of days of the
month) that the desired frequency is below the predicted monthly median MUF. This
probability is called Availability. W6ELProp puts Availability into four categories, and
the definitions of the limits for each category are at the bottom of the page. For example,
the 28.3MHz column at 1530 UTC shows a C. This means 10m should be open on 8 to
GD\VRIWKHPRQWKWR7RUHLWHUDWHZHGRQ¶WNQRZZKLFKVSHFLILFGD\V
these will be. From 1600 to 1900 UTC the probability increases to 16 to 24 days of the
month (B = 50% to 75%), with a signal level of about S8.

7KDW¶VLWIRUDEDVLFLQWURGXFWLRQWR:(/3URS7KHEHVWWKLQJWRGRLVWRGLYHLQDQGUXQ
predictions ±,¶PVXUH\RX¶OOEHFRPHIDPLOLDUZLWKLW)RUPRUHGHWDLOVJRWR+HOSand
WKHQ&RQWHQWV7KHUH¶VORWVRIJRRGLQIRUPDWLRQWKHUH+DYHIXQSOD\LQJZLWKDOOWKH
features of W6ELProp. The end result should be more DX success. Good luck!

One last comment - check out the mapping feature (under the Map menu) ±LW¶VUHDOO\
great for visualizing low band openings, where the dark ionosphere is the best place to be
for our RF.