[CQ-Contest] Radio Laws of Propagation....Have they been re-invented?

Fri Feb 28 09:22:25 EST 2014

I don't have any comment on KP4KE/NP2P, but I would like to make some comments re using the RBN data to compare stations:

The procedure of picking two SINGLE skimmer spots at similar times is never a reliable way to compare signal strengths. Remember, different stations, even if they are spotted at nearly the same time, are at different frequencies which may have very different QRM, QSB, etc. Skimmer can measure properties of a signal, but there is a large fluctuation. This fluctuation is not necessarily just from propagation differences. I expect the signal strength reported already includes some averaging over the time skimmer has decoded the callsign, but more is really needed for accurate strength measurements.

A good example is the skimmer measurement of phase and amplitude used for I/Q balance correction (which of course has no propagation difference issues). The I/Q balance measurements is only possible through averaging a very large number of signal measurements. In my case I wrote my own code for SO2SDR to do a similar averaging so I am quite aware of the amount of fluctuation- it can be huge, even for strong signals! Understanding that you have to use the data in a statistical sense is important- single measurements are not very reliable.

To get a meaningful comparison, you need to average a series of skimmer spots occurring in a short time (< 1 hour?) together. This is most easily seen using the "Spot analysis tool" on the reversebeacon web page. You want to look at the average of the measurements over a certain time, not the peaks or dips.

To give a concrete example from the K1LZ/N1UR comparison, look up 02/15/2014, choose the DF4UE skimmer:

10m : around 1430z: K1LZ fluctuates around about 40 dB, N1UR around 30 dB

15m: draw two horizontal lines through the points between 1500 and 1800. K1LZ is about 7dB stronger. If you looked at the K1LZ outlier point at 1549z and compared to N1UR at 1547z, you might think N1UR is much louder.

20,40: similar comparisons can be made, showing 5-10 dB in favor of K1LZ.

80,160: not enough data here

So yes, N1UR may have a good qth, but I think skimmer data does show that K1LZ is nearly 10 dB louder into Europe in this case on all bands.

Tor
N4OGW/5

________________________________
 From: Edward Sawyer <SawyerEd at Earthlink.net>
To: cq-contest at contesting.com 
Sent: Thursday, February 27, 2014 2:51 PM
Subject: Re: [CQ-Contest] Radio Laws of Propagation....Have they been re-invented?

I took great interest in Herb's post.  As a competitive Low Power
competitor, this issue is very important to me.  I thought I would share my
observations.

I routinely use RBN history after a CW contest to compare my station to the
top stations in W1.  This time I did so with K1LZ (arguably one of the top
stations on the East Coast).  I am trusting (and know those guys) that they
are using 1500W output.  I am using 150W output.  So apples to apples, they
have a 10dB advantage on me in EU (where I grabbed this data).  I have found
that the RBN can be deceiving for good scientific comparison.  Ignore the
lines that connect the dots.  The dots are the only actual datapoint.  Also,
I only use data that is +/- 2 to 3 mins maximum for comparison.  You can
easily see the same station vary by well over 10dB in the span of an hour so
if the points are really not close - you are kidding yourself on comparison.
Often some points are literally at the same clock minute or within 1 minute.
This is good data I feel.  There were 4 good points to compare in this way
on Feb 15 between K1LZ and N1UR.  Here is the data:

- 40M  0343Z K1LZ +11dB -  net of power +1dB

- 40M 2138Z K1LZ +13dB - net of power +3dB

- 20M 1901Z K1LZ +6dB - net of power -4dB

- 10M 1426Z K1LZ +10dB - net of power 0dB

In these instances only, K1LZ antennas are beating me +1 - +3dB on 40M,
losing to mine by 4dB on 20M, and equal to mine on 10M.  Note that in no
case was my 150W signal ever equal or exceeding K1LZ using 1500W.  K1LZ uses
3/3 on 40M, 5/5/5 on 20M, and 7/7/7 on 10M all on 150 ft towers but I
believe on pretty flat terrain.  N1UR is using 3 on 40M @85ft, 4/4 at 85 and
35ft on 20M, and an 8el at 30ft on a 45 ft boom on 10M.  I have
exceptionally enhanced terrain to EU that is clear in these numbers -
elevation 1200ft, with a 100 ft ridge drop off within 75ft of the tower
centerpoint and then rolling slope down to 500ft for about 2 miles in the EU
direction - yes I picked the location for DX contesting.  The data would
suggest (since I assume that K1LZ is using the same low loss hardline that I
am - or better) that my terrain is giving me a 2 - 6dB advantage assuming
typical antenna design gain comparisons.  This of course will vary contest
to contest based on conditions.

Now lets compare the exact same look using the same datapoint only within
+/- 3mins max between KP2KE and NP2P (using NA receivers).  There were also
a couple of exact time datapoints.

- 20M 2314Z KP4KE +13dB

- 40M 0526Z KP4KE +4dB

- 40M 0834Z KP4KE +3dB

- 40M 1059Z KP4KE +2dB

- 80M 0556Z KP4KE +2dB

- 80M 0727Z KP4KE +5dB

- 80M 0900Z KP4KE +4dB

- 160M 0415Z KP4KE + 5dB

Note that in no case is NP2P stronger or equal to KP4KE.  If you assume that
KP4KE was using 150W (the low power limit) regardless of exchange or 3830
post, then NP2P should have a 10dB signal advantage on him - all else being
equal.  Since the antennas are described as virtually identical in Herb's
post (on 40 and 80M), then only terrain/conductivity (near the sea) and
losses in feedlines could be left (other than power).  These combined
advantages of terrain, conductivity, and coax losses of KP4KE would have to
range from +12 - +20dB in KP4KE's favor given the data presented.  Is it
possible?  Potentially, but you would have to look at the two locations and
it would have the immediate reaction of "now it may make sense" to be
plausible.  In my opinion.

Are Herb's observations reasonable and worthy?  From this data, absolutely.
Does it prove cheating?  Possibly, if there is no other reasonable
explanation and the results occur contest after contest - in my opinion.

FYI - I have had more than one curious contester come check my place out
because I am pretty darned loud in EU for a low power station.  I own no
amplifiers and don't even have room for them in the operating area.  They
see the aluminum and gaze at the clear shot at high sloping elevation to UA9
- G-DL-I-9A-4X and Africa and kind of uncontrollably nod their head a few
times.  The view and hardware make it possible and plausible.  I think any
booming low power contester certainly - or any winning contester really -
should be prepared for reasonable questions and should have ready answers.
After all - physics doesn't play favorites - but it can be manipulated in
your favor..

Lastly, I note that KP4KE has not yet uploaded his log to ARRL to formally
declare his entry class.

73

Ed  N1UR

_______________________________________________
CQ-Contest mailing list
CQ-Contest at contesting.com
http://lists.contesting.com/mailman/listinfo/cq-contest