[TowerTalk] Re: Can takeoff angle be too low?

Bill Tippett btippett@alum.mit.edu
Wed, 20 Dec 2000 19:34:36 +0000


        Attached is a follow-up from Dean N6BV.  It's too bad
you can't see his attached files but here is a summary of the Europe
propagation statistics for GA and AL.  (Yes, I'm in NC but GA is the
closest QTH to me for which VOACAP statistics are available).

TOA (Degrees)      GA %                 AL % (% do not total 100 due to
1                   0                   10    visual interpolation errors)
2                   0                   11.5
3                   8                   14
4                  15                   10
5                  19                   11.5
6                   8                    9.5
7                  19                    6
8                   8                    7
9                  11                    7
10                 11                    4
11                  0                    0.5
12                  0                    2
13                  0                    0.5
14                  0                    1
15                  0                    0.5
16                  0                    0.5
17                  0                    2
18                  0                    0.5

Bottom line:  Georgia has a much tighter distribution of TOA's with Alabama
having both lower (<3 degrees) AND higher (>10 degrees) TOA's to Europe.
As many have said before, there is no substitute for modeling your own
specific QTH, terrain and antenna system using YT.  This data clearly
shows me that both of us are right for our specific locations...maybe
I'm just lucky that I can cover all propagation modes with a smaller
combination of antennas than would be required if I were in Alabama.

                                                73,  Bill  W4ZV

Date: Wed, 20 Dec 2000 09:03:51 -0800
From: "R. Dean Straw" <n6bv@arrl.org>
Subject: RE: [TowerTalk] Re: Can takeoff angle be too low?
To: n4kg@juno.com
Cc: btippett@alum.mit.edu

Tom/Bill:

Thanks for copying me on this discussion.

> We're not that far apart.  Your low stack covers 4 to 12 degrees,
> being superior to your full stack from 8 to 12 degrees.

N6BV: See attached 10m-GA.PCX file, which represents a screen shot from the
YT program. The elevation-angle statistics are for Atlanta, GA, to all of
Europe on 10 meters. (I don't have statistics for North Carolina -- don't
you live in NC, Bill? However GA should be reasonably close to what happens
in NC.) It should be abundantly clear that the high stack of four antennas
is way too high for the roughly 2 x 11.5% = 23% of time when the band is
open above 9 degrees in takeoff angle.

N6BV: The four-stack will, however, be superior to a 70/35' stack for
elevation angles less than about 8 degrees. The four-stack will be superior
to a single 35' Yagi below the crossover point of about 9 degrees. At 10
degrees, for example, the single 35' Yagi will be some 2.5 dB better than
the four-stack. Note however that the 70/35' stack will be some 3.5 dB
better at 10 degrees than the four-stack. As I said, for much of the time,
the four-stack would be simply too high, at least from Georgia.

N6BV: Now, let's look at the VOACAP predictions for 10 meters for November
with an SSN = 100 (about where we're at now), using isotropic antennas at
both ends of the circuit, but with gain on the transmitting and receiving
ends of 12 dBi. This is about the peak gain you'd get for a triband
horizontal Yagi over flat ground. (FYI, VOACAP requires a "workaround" for
the TX end of using more power than normal, while on the RX end it allows
you to specify extra gain beyond isotropic. Go figure.)

N6BV: I'm attaching a snippet of the huge Method 25 printout for 15 UTC from
Raleigh, NC (is that near you, Bill?) and Bonn, Germany.

 SUMMARY    4 MODES   FREQ =  28.4 MHZ  UT =  15.0
                                                      Most REL
                 2.F2      3.F2      3.F2      3. E       2.F2
  TIME DEL.     23.17     24.03     24.15     22.81      23.17
  ANGLE          1.32     11.37     12.07      1.24       1.32
  VIR. HITE    271.01    340.80    356.52    125.30     271.01
  TRAN.LOSS    137.68    156.67    159.47   1013.64     137.68
  T. GAIN       12.00     12.00     12.00     12.00      12.00
  R. GAIN       12.00     12.00     12.00     12.00      12.00
  ABSORB         2.88      1.95      1.88      2.89
  FS. LOSS     138.35    138.67    138.71    138.21
  FIELD ST.     18.35     -0.64     -3.44   -857.61      18.43
  SIG. POW.   -105.92   -124.91   -127.71   -981.88    -105.84
  SNR           70.83     51.84     49.04   -805.13      70.91
  MODE PROB      0.80      0.44      0.44      0.00       0.80
  R. PWRG     1000.00   1000.00   1000.00   1000.00      28.87
  RELIABIL       0.41      0.13      0.12      0.00       0.42
  SERV PROB      0.20      0.08      0.07      0.00       0.20
  SIG LOW       25.00     25.00     25.00     14.19      25.00
  SIG  UP       11.06     23.62     25.00      7.68      12.42
  NOISE =   -177      S. POWER = -105.8
  SIGNAL =   14.2   12.2    7.7  /     3.2     4.2     1.7
  NOISE =     9.6 -176.7    5.7  /     1.4     5.1     1.6
  RELIAB =   13.7   70.9   26.8
  SPROB =    34.2   44.0   34.2

N6BV: Yes, this is somewhat overwhelming, but let's wade through, for fun
and edification. Note that the "T. GAIN" and "R. GAIN" are 12 dB, just what
I said above, for all modes VOACAP found. There appear to be three viable
modes: a 2F2 hop at a very low takeoff angle of 1.32 degrees, resulting in a
received signal power of -105.92 dBW and an SNR of 70.83 dB/1Hz bandwidth.

N6BV: There is a 3F2 hop at a takeoff of 11.32 degrees, resulting in a
signal level of -124.91 dBW, about 20 dB weaker than the 2F2 mode (at least
for isotropic antennas). The second 3F2 hop is the Pedersen mode at 12.07
degrees takeoff angle. It results in a received signal of -127.71 dB, about
3 dB down from the main 3F2 mode, and 22 dB down from the 2F2 mode signal.
Note that time delay for the two 3F2 hops (modes) is very close to each
other, as you'd expect, given the close elevation angles.

N6BV: If these two 3F2 signals get rotated out of phase in the vagaries of
the ionosphere and since they are close together in amplitude, there's a
good chance that they'd cause severe fading -- if the antenna used at the
receiving side were low in height and thence tended to reject the dominant
2F2 mode. Note that this is pretty close to the situation we'd expect over
flat ground for a single 35' Yagi compared to the stack of four. At 2
degrees (I can't read the 1.32 degree angle that well...), the 35' antenna
is down some 13.5 dB to the four-stack. Of course, there would also be
fading due to combinations of the 2F2 and both 3F2 modes, no matter which
antenna system is being used.

> Back in 1980 when MUF's were especially high, I estimated
> that Europeans were peaking at 10 to 15 degrees during
> mid day based on observations on my 40 ft high 5L10 vs 80 ft
> high 2L Quad with a NULL from 11 to 15 degrees and a flat
> terrain towards Europe.  MUF's in subsequent maximums
> have not reached the levels of the 1980 era.

N6BV: Now, just for fun, look at 10m-ALA.PCX, which overlays the elevation
statistics for Alabama on the same graph. Things are *much* more complicated
for Alabama (Montgomery) than for Atlanta, GA. The spread of angles on 10
meters into Europe from Alabama goes as high as 18 degrees. Here, there is
some possibility that the low 35-foot antenna would have times when it is
superior to the other antenna combinations.

N6BV: Really low elevation angles come into play from Alabama much more
often than from Georgia, statistically speaking, of course. The angle is as
low as 1 degree some 10% of the times when the 10-meter band is open to
Europe.

N6BV: Here's a snippet of the VOACAP Method 25 output for 15 UTC from
Montgomery, AL:

 SUMMARY    2 MODES   FREQ =  28.4 MHZ  UT =  15.0
                                  Most REL
                 3.F2      3.F2       3.F2
  TIME DEL.     26.22     27.23      26.22
  ANGLE          7.27     12.82       7.27
  VIR. HITE    293.65    432.50     293.65
  TRAN.LOSS    148.38    170.33     148.38
  T. GAIN       12.00     12.00      12.00
  R. GAIN       12.00     12.00      12.00
  ABSORB         2.27      1.74
  FS. LOSS     139.42    139.75
  FIELD ST.      7.65    -14.30       7.67
  SIG. POW.   -116.62   -138.57    -116.59
  SNR           60.13     38.18      60.15
  MODE PROB      0.57      0.57       0.57
  R. PWRG     1000.00   1000.00      39.62
  RELIABIL       0.17      0.04       0.17
  SERV PROB      0.12      0.03       0.12
  SIG LOW       25.00     25.00      25.00
  SIG  UP       16.33     25.00      16.50
  NOISE =   -177      S. POWER = -116.6
  SIGNAL =   13.4   11.4    7.1  /     2.9     3.8     1.4
  NOISE =     9.6 -176.7    5.7  /     1.4     5.1     1.6
  RELIAB =   17.5   60.2   26.8
  SPROB =    33.9   33.3   33.9

N6BV: Here, there are only two modes available at 15 UTC. The Pedersen ray
is at 12.82 degrees, with a signal strength of -138.57 dBW, while the main
3F2 mode comes in at 7.27 degrees, with a signal strength of -116.62 dBW.
This is almost 22 dB stronger. It's pretty likely that a 35' high Yagi
wouldn't even be able to detect the Pedersen ray, even though its response
at 12.82 degrees would be optimal. The Pedersen ray is simply too weak, with
a received SNR of only 38.18 dB/1Hz.

N6BV: So, in Alabama anyway, in November with a Very High (but not Ultra
High) level of solar activity, the higher antennas would definitely rule the
roost. So, you ask, what happens when the solar activity is much higher, say
SSN = 150?

 SUMMARY    4 MODES   FREQ =  28.4 MHZ  UT =  15.0
                                                      Most REL
                 3.F2      4.F2      4.F2      4. E       3.F2
  TIME DEL.     26.08     26.91     28.35     25.46      26.08
  ANGLE          6.34     13.68     20.00      3.31       6.34
  VIR. HITE    271.52    310.73    437.09    125.30     271.52
  TRAN.LOSS    142.16    156.63    178.81   1264.04     142.16
  T. GAIN       12.00     12.00     12.00     12.00      12.00
  R. GAIN       12.00     12.00     12.00     12.00      12.00
  ABSORB         2.88      2.04      1.56      3.22
  FS. LOSS     139.38    139.65    140.10    139.17
  FIELD ST.     13.87     -0.60    -22.78  -1108.01      14.02
  SIG. POW.   -110.40   -124.87   -147.05  -1232.28    -110.24
  SNR           66.35     51.88     29.70  -1055.53      66.50
  MODE PROB      0.77      0.45      0.45      0.00       0.77
  R. PWRG     1000.00   1000.00   1000.00   1000.00      33.27
  RELIABIL       0.23      0.07      0.02      0.00       0.24
  SERV PROB      0.16      0.08      0.02      0.00       0.16
  SIG LOW       25.00     25.00     25.00     13.42      25.00
  SIG  UP        9.82     17.13     25.00      7.08      10.46
  NOISE =   -177      S. POWER = -110.2
  SIGNAL =   13.4   11.4    7.1  /     2.9     3.8     1.4
  NOISE =     9.6 -176.7    5.7  /     1.4     5.1     1.6
  RELIAB =   11.9   66.5   26.8
  SPROB =    33.9   39.5   33.9

N6BV: Now things are a little more interesting into Alabama, with this Ultra
High level of solar activity. Here we have two modes that yield workable
SNRs (that is, greater than 43 dB for SSB): 3F2 and 4F2, at launch angles of
6.34 and 13.68 degrees respectively. The 4F2 Pedersen at 20.0 degrees is
pretty much down in the mud.

N6BV: The two viable modes yield received signal levels of -110.40 dBW
and -124.87 dBW, a 14.47 dB difference. A single 35' high Yagi (as opposed
to the isotropics assumed in this VOACAP model) would be optimal for the 4F2
mode, and would be down only 4.5 dB in response for the 3F2 mode. This means
for real 35' high antennas that the 3F2 mode would still dominate: at 2 x
4.5 = 9 dB down, narrowing the 14.47 dB gap for isotropic antennas to 5.47
dB for real 35' Yagis at both ends of the circuit.

> Back then, N4AR (K4GSU) installed a 4 high stack on
> 10M at 35 / 65 / 95 / 125 ft.  He noticed considerable
> echo on Europeans which I attribute to the high gain
> he had at 5 or 6 degrees which dropped off above 10
> degrees whereas most Europeans were putting more
> energy into the higher angle modes.

N6BV: The "echos" reported for high stacks are interesting phenomena, not
explained directly by the VOACAP modeling above. I believe that these echos
are mostly due to other propagation modes, mainly involving off-azimuth
scatter mechanisms that introduce much longer propagation paths and hence
longer propagation delays, resulting in the echos. For example, the
well-known scatter path from the Northeast USA to Europe when 10 meters
direct-path is dead during low solar activity is the one bouncing signals
off the Azores. This is found when the antennas in the USA are aimed at 90
degrees, rather than directly at 45 degrees. The high, big stacks have
enough gain to make these weaker scatter paths work when lower antennas
can't hack it, even during high levels of solar activity.

> When he put up a 4L10 at 40 ft, the echo disappeared
> and the Europeans were actually stronger than on his
> 4 high stack.

N6BV: Unless I could see the exact terrain involved for all these antennas,
I couldn't comment on this anecdotal evidence! However, his four-stack had a
huge null at about 12-18 degrees. Anything signal falling into this "hole"
would be gone.

> I also observed that locals with tribanders at 80 to
> 100 ft were 'last in line' in 10M pileups once the band
> was wide open to Europe.

This is not terribly surprising. See the file 80Ft-AL.PCX attached here.
Assuming that the locals you refer to are on flat land, and assuming that
the incoming angles were relatively high, say 12 degrees, the 80-foot
antenna would be at a severe disadvantage compared to a 35' one, being down
about 16 dB because of the null. So what's new?!

> I expect that the minimum path loss does occur at
> low angles BUT this means that stations at BOTH
> ends must have antennas which favor those angles.
> Most Europeans do NOT.  I expect THEIR signals
> will peak at higher angles since they are transmitting
> more energy at the higher angles that are still supported
> by the ionosphere.  Perhaps I didn't state this clearly.
> That is the point I was trying to make regarding
> the monitoring of HF Broadcast stations vs. the average
> EU amateur station with low antennas.

N6BV: Arghhhh... The height of the antennas at either end of the circuit
does *not* change what modes the ionosphere dictates, only what the actual
received signal strengths are.

> I have often wondered if there is some "bending" in
> the E layer, which may be more prevalent in the
> southern states, especially on 20M in the early
> afternoon.

N6BV: By "bending" are you referring perhaps to off-azimuth modes?

N6BV: Tom, I urge you to try these programs yourself -- VOACAP and YT. There
is a lot of good science that has been done by the US Government over the
last 40 years in IONCAP/VOACAP.

N6BV: Again, if you or Bill wish to re-post this on the reflector, be my
guest. Please post it all, however. That's only fair!

73, Dean, N6BV

P.S.  See ARRL Antenna Compendium 6 and
http://www.arrl.org/notes/antbook/yt-files.htm
for more information on the revisions Dean made to his original VOACAP
assumptions (using non-isotropic antennas)...de W4ZV


--
FAQ on WWW:               http://www.contesting.com/FAQ/towertalk
Submissions:              towertalk@contesting.com
Administrative requests:  towertalk-REQUEST@contesting.com
Problems:                 owner-towertalk@contesting.com