Need K3 dB Measurements on new 160M Antenna

[Jim Brown-10]

Crew,

I've put up a pair of new antennas for 160M that are predicted to have
about 3dB of gain over my existing vertical. Without going into a lot of
detail, each antenna is a wire sloping off of my 120 ft tower, one going
east, one going west. The tower acts as a reflector. The tower and each
wire have 4 elevated radials. The antenna is working -- I've made four QSOs
with VK, one with FK8, and one with FO in about two hours -- but I need to
figure out if it's working better than my existing omni vertical.

By doing a lot of listening, I can clearly confirm the predicted 6dB or so
of front to back, but QSB makes it hard to get a handle on gain, and I'm
only expecting 2-3dB. That's where the dB meter in the K3 comes in handy.

My method is to key down on one antenna for a while, let you get a reading,
switch to the other and do the same, then back and forth between them,
again for long enough for you to get a good reading. There IS a lot of QSB
on the band, so you'll need to do a lot of mental averaging. Please let me
know, OFFF THE LIST, if this is something that you would like to help me
with. I need reports from stations that are between about 30 degrees
azimuth and 120 degrees azimuth of my QTH south of San Francisco. In other
words, my antenna is aimed at about 75 degrees (ENE), and I need reports
that are within 50 degrees of being on axis. I could also use measurements
from KH6 of the antenna that goes in that direction.

BTW -- to access the dB meter in the K3, you need to hit the Display button
once, then rotate the second VFO knob clockwise until you see a reading in
mV. Let the knob in that position for about 15 seconds, then rotate it one
more position clockwise. Now, the dB reading will be relative to whatever
the average IF output was when the knob was in the mV position. Note also
that the reading goes plus and minus.

73, Jim K9YC






______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[G3TCT]

Jim
Afraid I can't help with listening on 160, but an interesting topic, and
a measurement technique applicable to other situations. My observation
is that the mental averaging will be rather tricky, especially if the
QSB is slow and unpredictable. What we need is something (software?)
which will record the strength seen by the K3 over a minute or 3, and
then provide a statistical analysis. You would then get a picture of the
average, peak, trough and distribution of strengths.
I wonder if something like this exists already - or could be included in
P3 s/w?

Graham

On 19:59, Jim Brown wrote:

> Crew,
>
> I've put up a pair of new antennas for 160M that are predicted to have
> about 3dB of gain over my existing vertical. Without going into a lot of
> detail, each antenna is a wire sloping off of my 120 ft tower, one going
> east, one going west. The tower acts as a reflector. The tower and each
> wire have 4 elevated radials. The antenna is working -- I've made four QSOs
> with VK, one with FK8, and one with FO in about two hours -- but I need to
> figure out if it's working better than my existing omni vertical.
>
> By doing a lot of listening, I can clearly confirm the predicted 6dB or so
> of front to back, but QSB makes it hard to get a handle on gain, and I'm
> only expecting 2-3dB. That's where the dB meter in the K3 comes in handy.
>
> My method is to key down on one antenna for a while, let you get a reading,
> switch to the other and do the same, then back and forth between them,
> again for long enough for you to get a good reading. There IS a lot of QSB
> on the band, so you'll need to do a lot of mental averaging. Please let me
> know, OFFF THE LIST, if this is something that you would like to help me
> with. I need reports from stations that are between about 30 degrees
> azimuth and 120 degrees azimuth of my QTH south of San Francisco. In other
> words, my antenna is aimed at about 75 degrees (ENE), and I need reports
> that are within 50 degrees of being on axis. I could also use measurements
> from KH6 of the antenna that goes in that direction.
>
> BTW -- to access the dB meter in the K3, you need to hit the Display button
> once, then rotate the second VFO knob clockwise until you see a reading in
> mV. Let the knob in that position for about 15 seconds, then rotate it one
> more position clockwise. Now, the dB reading will be relative to whatever
> the average IF output was when the knob was in the mV position. Note also
> that the reading goes plus and minus.
>
> 73, Jim K9YC
>
>
>
>
>
>
>
>    

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[Dick Dickinson]

Jim -

Give www.w4mq.com a try!


73,
Dick - KA5KKT

----------------------------------------------------------------------------
----------------
On 19:59, Jim Brown wrote:
> Crew,
>
> I've put up a pair of new antennas for 160M that are predicted to have
> about 3dB of gain over my existing vertical. Without going into a lot of
> detail, each antenna is a wire sloping off of my 120 ft tower, one going
> east, one going west. The tower acts as a reflector. The tower and each
> wire have 4 elevated radials. The antenna is working -- I've made four
QSOs
> with VK, one with FK8, and one with FO in about two hours -- but I need to
> figure out if it's working better than my existing omni vertical.
>
> By doing a lot of listening, I can clearly confirm the predicted 6dB or so
> of front to back, but QSB makes it hard to get a handle on gain, and I'm
> only expecting 2-3dB. That's where the dB meter in the K3 comes in handy.
>
> My method is to key down on one antenna for a while, let you get a
reading,

> switch to the other and do the same, then back and forth between them,
> again for long enough for you to get a good reading. There IS a lot of QSB
> on the band, so you'll need to do a lot of mental averaging. Please let me
> know, OFFF THE LIST, if this is something that you would like to help me
> with. I need reports from stations that are between about 30 degrees
> azimuth and 120 degrees azimuth of my QTH south of San Francisco. In other
> words, my antenna is aimed at about 75 degrees (ENE), and I need reports
> that are within 50 degrees of being on axis. I could also use measurements
> from KH6 of the antenna that goes in that direction.
>
> BTW -- to access the dB meter in the K3, you need to hit the Display

button
> once, then rotate the second VFO knob clockwise until you see a reading in
> mV. Let the knob in that position for about 15 seconds, then rotate it one
> more position clockwise. Now, the dB reading will be relative to whatever
> the average IF output was when the knob was in the mV position. Note also
> that the reading goes plus and minus.
>
> 73, Jim K9YC

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[Guy, K2AV]

Alas, after years of doing A/B manual coax switch tests, I have given
up on that exposing anything better than 10 db differences on anything
other than stable local signals.  And that is suspect because it is
often ground wave, which bears no resemblance to sky wave.

I built a 60 hz switch which used 12VAC plus and negative to
alternately bias off left and right diodes connecting a common RF
output port to two RF input ports.  Being careful to use no AGC or
very slow AGC, this presented an audio from the RX which showed to an
oscilloscope the signal comparison between A and B that could be
measured on the scope and converted to dB.

I also quickly learned that I had no hope whatsoever of perceiving a
difference less than 3 dB in my ear and didn't do all that well with
less than 6.

I used that to compare signals on various antennas and showed it to
the owner.  But so ingrained is the idea of manual A/B coax switching
that he was back to judging results the old way, and discarding
methods that gained a dB here and there, because "he couldn't hear it,
and was going to trust his ears."

The main problem of the device was an unambiguous way of identifying
the port on the oscilloscope display.  I have an idea of using a PIC
device instead of the house AC to create the switching intervals, one
which starts a sequence with a "long" A port and ends with a long B
port and 8 regular ports in between, with a space between the two long
ports. That would always unambiguously identify the A and B signals.
Follow that with a program to analyze the audio levels and present
peaks, minimums and averages for both signals and signal-to-noise, and
you now have an antenna analyzer that can show you real differences
between antennas real-time.

To me anyway, that sounds like a tailor-made Elecraft gizmo kit.  I
think you could sell tens of thousands of those.  Really surprised
something like that not already around and part of during-contest
comparisons between antennas.

If the gizmo had the ability to decode the results and put it on a LED
display marked with port A on one end and port B on the other, with
the middle LED meaning equal, with two or three ranges, it would be
the cat's meow for comparing two antennas.

73, Guy.

On Sat, Aug 28, 2010 at 11:49 AM, Graham Kimbell G3TCT
<[hidden email]> wrote:

> Jim
> Afraid I can't help with listening on 160, but an interesting topic, and
> a measurement technique applicable to other situations. My observation
> is that the mental averaging will be rather tricky, especially if the
> QSB is slow and unpredictable. What we need is something (software?)
> which will record the strength seen by the K3 over a minute or 3, and
> then provide a statistical analysis. You would then get a picture of the
> average, peak, trough and distribution of strengths.
> I wonder if something like this exists already - or could be included in
> P3 s/w?
>
> Graham
>
> On 19:59, Jim Brown wrote:
>> Crew,
>>
>> I've put up a pair of new antennas for 160M that are predicted to have
>> about 3dB of gain over my existing vertical. Without going into a lot of
>> detail, each antenna is a wire sloping off of my 120 ft tower, one going
>> east, one going west. The tower acts as a reflector. The tower and each
>> wire have 4 elevated radials. The antenna is working -- I've made four QSOs
>> with VK, one with FK8, and one with FO in about two hours -- but I need to
>> figure out if it's working better than my existing omni vertical.
>>
>> By doing a lot of listening, I can clearly confirm the predicted 6dB or so
>> of front to back, but QSB makes it hard to get a handle on gain, and I'm
>> only expecting 2-3dB. That's where the dB meter in the K3 comes in handy.
>>
>> My method is to key down on one antenna for a while, let you get a reading,
>> switch to the other and do the same, then back and forth between them,
>> again for long enough for you to get a good reading. There IS a lot of QSB
>> on the band, so you'll need to do a lot of mental averaging. Please let me
>> know, OFFF THE LIST, if this is something that you would like to help me
>> with. I need reports from stations that are between about 30 degrees
>> azimuth and 120 degrees azimuth of my QTH south of San Francisco. In other
>> words, my antenna is aimed at about 75 degrees (ENE), and I need reports
>> that are within 50 degrees of being on axis. I could also use measurements
>> from KH6 of the antenna that goes in that direction.
>>
>> BTW -- to access the dB meter in the K3, you need to hit the Display button
>> once, then rotate the second VFO knob clockwise until you see a reading in
>> mV. Let the knob in that position for about 15 seconds, then rotate it one
>> more position clockwise. Now, the dB reading will be relative to whatever
>> the average IF output was when the knob was in the mV position. Note also
>> that the reading goes plus and minus.
>>
>> 73, Jim K9YC
>>
>>
>>
>>
>>
>>
>>
>>
>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
>

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[Jim Brown-10]

On Sat, 28 Aug 2010 13:38:59 -0400, Edward Dickinson, III wrote:

>Give www.w4mq.com a try!

Very slick!  Now, if it can hear me from 2,500 miles away. :)

73, Jim K9YC Santa Cruz, CA



______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[David Cutter]

I would try a chart recorder in this arrangement; there's no mistaking which
channel is which.  Of course use a modern version using pc sound card then
you can squash up or expand the results for examination.

David
G3UNA

> Alas, after years of doing A/B manual coax switch tests, I have given
> up on that exposing anything better than 10 db differences on anything
> other than stable local signals.  And that is suspect because it is
> often ground wave, which bears no resemblance to sky wave.
>
> I built a 60 hz switch which used 12VAC plus and negative to
> alternately bias off left and right diodes connecting a common RF
> output port to two RF input ports.  Being careful to use no AGC or
> very slow AGC, this presented an audio from the RX which showed to an
> oscilloscope the signal comparison between A and B that could be
> measured on the scope and converted to dB.
>
> I also quickly learned that I had no hope whatsoever of perceiving a
> difference less than 3 dB in my ear and didn't do all that well with
> less than 6.
>
> I used that to compare signals on various antennas and showed it to
> the owner.  But so ingrained is the idea of manual A/B coax switching
> that he was back to judging results the old way, and discarding
> methods that gained a dB here and there, because "he couldn't hear it,
> and was going to trust his ears."
>
> The main problem of the device was an unambiguous way of identifying
> the port on the oscilloscope display.  I have an idea of using a PIC
> device instead of the house AC to create the switching intervals, one
> which starts a sequence with a "long" A port and ends with a long B
> port and 8 regular ports in between, with a space between the two long
> ports. That would always unambiguously identify the A and B signals.
> Follow that with a program to analyze the audio levels and present
> peaks, minimums and averages for both signals and signal-to-noise, and
> you now have an antenna analyzer that can show you real differences
> between antennas real-time.
>
> To me anyway, that sounds like a tailor-made Elecraft gizmo kit.  I
> think you could sell tens of thousands of those.  Really surprised
> something like that not already around and part of during-contest
> comparisons between antennas.
>
> If the gizmo had the ability to decode the results and put it on a LED
> display marked with port A on one end and port B on the other, with
> the middle LED meaning equal, with two or three ranges, it would be
> the cat's meow for comparing two antennas.
>
> 73, Guy.
>
> On Sat, Aug 28, 2010 at 11:49 AM, Graham Kimbell G3TCT
> <[hidden email]> wrote:
>> Jim
>> Afraid I can't help with listening on 160, but an interesting topic, and
>> a measurement technique applicable to other situations. My observation
>> is that the mental averaging will be rather tricky, especially if the
>> QSB is slow and unpredictable. What we need is something (software?)
>> which will record the strength seen by the K3 over a minute or 3, and
>> then provide a statistical analysis. You would then get a picture of the
>> average, peak, trough and distribution of strengths.
>> I wonder if something like this exists already - or could be included in
>> P3 s/w?
>>
>> Graham
>>
>> On 19:59, Jim Brown wrote:
>>> Crew,
>>>
>>> I've put up a pair of new antennas for 160M that are predicted to have
>>> about 3dB of gain over my existing vertical. Without going into a lot of
>>> detail, each antenna is a wire sloping off of my 120 ft tower, one going
>>> east, one going west. The tower acts as a reflector. The tower and each
>>> wire have 4 elevated radials. The antenna is working -- I've made four
>>> QSOs
>>> with VK, one with FK8, and one with FO in about two hours -- but I need
>>> to
>>> figure out if it's working better than my existing omni vertical.
>>>
>>> By doing a lot of listening, I can clearly confirm the predicted 6dB or
>>> so
>>> of front to back, but QSB makes it hard to get a handle on gain, and I'm
>>> only expecting 2-3dB. That's where the dB meter in the K3 comes in
>>> handy.
>>>
>>> My method is to key down on one antenna for a while, let you get a
>>> reading,
>>> switch to the other and do the same, then back and forth between them,
>>> again for long enough for you to get a good reading. There IS a lot of
>>> QSB
>>> on the band, so you'll need to do a lot of mental averaging. Please let
>>> me
>>> know, OFFF THE LIST, if this is something that you would like to help me
>>> with. I need reports from stations that are between about 30 degrees
>>> azimuth and 120 degrees azimuth of my QTH south of San Francisco. In
>>> other
>>> words, my antenna is aimed at about 75 degrees (ENE), and I need reports
>>> that are within 50 degrees of being on axis. I could also use
>>> measurements
>>> from KH6 of the antenna that goes in that direction.
>>>
>>> BTW -- to access the dB meter in the K3, you need to hit the Display
>>> button
>>> once, then rotate the second VFO knob clockwise until you see a reading
>>> in
>>> mV. Let the knob in that position for about 15 seconds, then rotate it
>>> one
>>> more position clockwise. Now, the dB reading will be relative to
>>> whatever
>>> the average IF output was when the knob was in the mV position. Note
>>> also
>>> that the reading goes plus and minus.
>>>
>>> 73, Jim K9YC
>>>
>>>

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[David Gilbert]

Why not just use the two receivers in the K3 ... one using the main
antenna port and the other using the RX antenna antenna port.  Take the
output from the LINE OUT port and feed it into a computer sound card to
record the resultant audio in stereo.  You can then compare anything you
want by looking at the waveforms side by side in an audio editor
(Audacity is free) with a db scale.  The comparison will be almost
instantaneous, and if you use a tone for the signal you can take out the
noise using a bandpass filter in the editor.  If you don't trust the
receivers to be identical, run the test both ways.   If you run the
recorded file through Spectrogram you can expand the db scale for
whatever precision you want.

73,
Dave   AB7E

>> Alas, after years of doing A/B manual coax switch tests, I have given
>> up on that exposing anything better than 10 db differences on anything
>> other than stable local signals.  And that is suspect because it is
>> often ground wave, which bears no resemblance to sky wave.
>>
>> I built a 60 hz switch which used 12VAC plus and negative to
>> alternately bias off left and right diodes connecting a common RF
>> output port to two RF input ports.  Being careful to use no AGC or
>> very slow AGC, this presented an audio from the RX which showed to an
>> oscilloscope the signal comparison between A and B that could be
>> measured on the scope and converted to dB.
>>
>> I also quickly learned that I had no hope whatsoever of perceiving a
>> difference less than 3 dB in my ear and didn't do all that well with
>> less than 6.
>>
>> I used that to compare signals on various antennas and showed it to
>> the owner.  But so ingrained is the idea of manual A/B coax switching
>> that he was back to judging results the old way, and discarding
>> methods that gained a dB here and there, because "he couldn't hear it,
>> and was going to trust his ears."
>>
>> The main problem of the device was an unambiguous way of identifying
>> the port on the oscilloscope display.  I have an idea of using a PIC
>> device instead of the house AC to create the switching intervals, one
>> which starts a sequence with a "long" A port and ends with a long B
>> port and 8 regular ports in between, with a space between the two long
>> ports. That would always unambiguously identify the A and B signals.
>> Follow that with a program to analyze the audio levels and present
>> peaks, minimums and averages for both signals and signal-to-noise, and
>> you now have an antenna analyzer that can show you real differences
>> between antennas real-time.
>>
>> To me anyway, that sounds like a tailor-made Elecraft gizmo kit.  I
>> think you could sell tens of thousands of those.  Really surprised
>> something like that not already around and part of during-contest
>> comparisons between antennas.
>>
>> If the gizmo had the ability to decode the results and put it on a LED
>> display marked with port A on one end and port B on the other, with
>> the middle LED meaning equal, with two or three ranges, it would be
>> the cat's meow for comparing two antennas.
>>
>> 73, Guy.

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[Geoffrey Mackenzie-Kennedy]

David,

The question remains IMHO whether the data taken from short term "on-air"
tests over distances which involve ionospheric propagation is useful in the
first place.

Unless the vertical patterns of the A and B antennas are very similar in
shape and lobe take off angle, it can be seen that changes in the angle of
arrival of the test signal would introduce errors to the results of "on-air"
tests.

I have no experience with 160m, but on 40m for example the change in the
angle of arrival of short path signals from the East Coast and Midwest can
be large, and can change rapidly. Sometimes my dipole 70ft AGL which is
end-on to the States, a "cloud`warmer" in that direction, will outperform a
40m beam pointed at the States - a disturbing result if taken to be the true
performance of the beam.

IMHO "on air " antenna tests need to be repeated many times over a fairly
long period, preferably with the same stations.

73,
Geoff
GM4ESD


David Cutter wrote on Sunday, August 29, 2010 6:36 AM

>I would try a chart recorder in this arrangement; there's no mistaking
>which
> channel is which.  Of course use a modern version using pc sound card then
> you can squash up or expand the results for examination.
>
> David
> G3UNA

>> Alas, after years of doing A/B manual coax switch tests, I have given
>> up on that exposing anything better than 10 db differences on anything
>> other than stable local signals.  And that is suspect because it is
>> often ground wave, which bears no resemblance to sky wave.




______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[David Cutter]

Geoff

I was in eQSO with L B Cebik some while ago on a similar tack; basically I
asked how do we test a 4 square design made in EZNEC, say, with on air
experience.  He said more or less what you've just said, ie long term
testing over wide azimuth and elevation.  I had just designed a passive
array for 80m with advantages over a 4 square.
73
David
G3UNA

> David,
>
> The question remains IMHO whether the data taken from short term "on-air"
> tests over distances which involve ionospheric propagation is useful in
> the first place.
>
> Unless the vertical patterns of the A and B antennas are very similar in
> shape and lobe take off angle, it can be seen that changes in the angle of
> arrival of the test signal would introduce errors to the results of
> "on-air" tests.
>
> I have no experience with 160m, but on 40m for example the change in the
> angle of arrival of short path signals from the East Coast and Midwest can
> be large, and can change rapidly. Sometimes my dipole 70ft AGL which is
> end-on to the States, a "cloud`warmer" in that direction, will outperform
> a 40m beam pointed at the States - a disturbing result if taken to be the
> true performance of the beam.
>
> IMHO "on air " antenna tests need to be repeated many times over a fairly
> long period, preferably with the same stations.
>
> 73,
> Geoff
> GM4ESD
>
>
> David Cutter wrote on Sunday, August 29, 2010 6:36 AM
>
>>I would try a chart recorder in this arrangement; there's no mistaking
>>which
>> channel is which.  Of course use a modern version using pc sound card
>> then
>> you can squash up or expand the results for examination.
>>
>> David
>> G3UNA
>
>>> Alas, after years of doing A/B manual coax switch tests, I have given
>>> up on that exposing anything better than 10 db differences on anything
>>> other than stable local signals.  And that is suspect because it is
>>> often ground wave, which bears no resemblance to sky wave.
>

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[Don Nelson]

Geoff,

To do good antenna testing you need an anechoic chamber or better
operate in deep space. My primary interest in radio are the microwave
bands. I do set up antenna ranges for 10 GHz (3cm), In doing so, I find
the wind blowing tree leaves will have a non trivial affect on the
observed signal even though the trees are "out of the way." I find that
I need to place the reference antenna and the antenna under test in
exactly the same place (x y and z position) to get true A/B comparisons.
This is demonstrated by taking two known antenna and see how repeatable
the observations can be.

Scaling these realizations up to HF antenna causes me to throw up my
hands. Yes we do the best we can testing antenna, but we have to be
realistic about the data we collect. When anyone makes claims about
their antenna I listen carefully for the details on how the antenna was
tested.

Maybe we should ask ARRL to fund a synchronous satellite over the US
with a beacon on each of the HF bands so that we can point our antenna
up and make more valid comparisons.

Don, N0YE





On 8/29/2010 5:22 AM, Geoffrey Mackenzie-Kennedy wrote:

> David,
>
> The question remains IMHO whether the data taken from short term "on-air"
> tests over distances which involve ionospheric propagation is useful in the
> first place.
>
> Unless the vertical patterns of the A and B antennas are very similar in
> shape and lobe take off angle, it can be seen that changes in the angle of
> arrival of the test signal would introduce errors to the results of "on-air"
> tests.
>
> I have no experience with 160m, but on 40m for example the change in the
> angle of arrival of short path signals from the East Coast and Midwest can
> be large, and can change rapidly. Sometimes my dipole 70ft AGL which is
> end-on to the States, a "cloud`warmer" in that direction, will outperform a
> 40m beam pointed at the States - a disturbing result if taken to be the true
> performance of the beam.
>
> IMHO "on air " antenna tests need to be repeated many times over a fairly
> long period, preferably with the same stations.
>
> 73,
> Geoff
> GM4ESD
>
>
> David Cutter wrote on Sunday, August 29, 2010 6:36 AM
>
>> I would try a chart recorder in this arrangement; there's no mistaking
>> which
>> channel is which.  Of course use a modern version using pc sound card then
>> you can squash up or expand the results for examination.
>>
>> David
>> G3UNA
>>> Alas, after years of doing A/B manual coax switch tests, I have given
>>> up on that exposing anything better than 10 db differences on anything
>>> other than stable local signals.  And that is suspect because it is
>>> often ground wave, which bears no resemblance to sky wave.
>
>
>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
>

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[Jim Brown-10]

On Sun, 29 Aug 2010 12:22:22 +0100, Geoffrey Mackenzie-Kennedy wrote:

>The question remains IMHO whether the data taken from short term "on-air"
>tests over distances which involve ionospheric propagation is useful in the
>first place.

Clearly any such testing requires the averaging of a large number of samples,
and careful comparative tests. That is exactly what I am trying to do -- a
large number of samples, carefully taken and carefully averaged.

For exactly that reason, I am primarily looking for GROUND WAVE tests when
practical. When there are lots of stations on 160M during the daylight hours
(which happens during a major contest), I can regularly work stations over
distances of 800 miles or more if they have good RX antennas and are running
legal power. For example, from my QTH 70 miles S of San Francisco, I can work
stations near Seattle (750 miles), Salt Lake City (600 miles), Phoenix (600
miles), and Denver (950 miles). While there is SOME fading of ground wave, it
is far less than sky wave.

>Unless the vertical patterns of the A and B antennas are very similar in
>shape and lobe take off angle, it can be seen that changes in the angle of
>arrival of the test signal would introduce errors to the results of "on-air"
>tests.

The vertical patterns predicted by NEC for the new antenna and the reference
antenna in the favored direction ARE nearly identical.

73, Jim K9YC


______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[Alan Bloom]

In a previous life I was a radio operator at W1AW for several years.  I
think it was about 1974 when the new 120-foot tower was installed.  We
wanted to compare the big (700 feet long as I recall) rhombic we had
previously used on 20 meters for the bulletin and code practice
transmissions with the new stacked Yagis on the 120-foot tower.  Both
antennas were pointed due west from Newington, CT.

So for a week or so a special CW bulletin was sent asking for reports of
the relative signal strength of the two antennas.  CW bulletins are sent
several times a day so we covered the different propagation conditions.
The test consisted of several iterations of sending "antenna A" or
"antenna B" followed by several seconds of key-down.  The antenna
assignment changed randomly between bulletins so people wouldn't know
which was which.  Listeners then sent in QSL cards with their signal
reports.

I was assigned the task of analyzing the data.  There was some variation
among stations, but we had enough reports to get consistent results.  I
plotted the relative signal strengths on a map of the United States.

The result was that, right on the bore-sight of the antennas (the
great-circle path from Newington to southern California) the two
antennas were just about identical.  However, as you moved up the
Pacific coastline, the stacked beams started to do better than the
rhombic.  By the time you got up to Oregon and Washington state, the
beams were consistently an S-unit or two better.

It makes sense.  For a given gain, stacked Yagis have a much broader
horizontal beamwidth than a rhombic.  It was nice to see theory
confirmed by measurement.

Al N1AL



On Sun, 2010-08-29 at 10:49 -0700, Jim Brown wrote:
> On Sun, 29 Aug 2010 12:22:22 +0100, Geoffrey Mackenzie-Kennedy wrote:
>
> >The question remains IMHO whether the data taken from short term "on-air"
> >tests over distances which involve ionospheric propagation is useful in the
> >first place.
>
> Clearly any such testing requires the averaging of a large number of samples,
> and careful comparative tests. That is exactly what I am trying to do -- a
> large number of samples, carefully taken and carefully averaged.


______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[Jack Smith-6]

  In evaluating diversity gain of a Z1501D active antenna against
various transmitting antennas, I use two spectrum analyzers, one
connected to the two antennas to be evaluated. The spectrum analyzers
are set for 0 Hz span and a 100 second sweep, providing a signal
strength data point every 100 ms.

Under software control over the IEEE-488 bus, both spectrum analyzers
start sweeps at a synchronized time. The signal strength data from each
is saved to the controlling computer hard drive. The sweeps are repeated
and data collected for a reasonable time, usually 24 hours per frequency.

The data is then post processed to derive useful statistics, such as the
probability that antenna A provides a stronger signal than antenna B as
a percentage of time, in the form of signals from "Antenna A are X dB
stronger (or weaker) than Antenna B Z percent of the time."

Of course, other interesting things can be observed from the data, such
as fading statistics - is it Rician or Rayleigh fading, for example.

Writing this up is on my list of things to get to, but there are many
things ahead of it in the list.

A short answer is that in fact even modest separation (100 feet or less)
between the broadband active antenna and a wire antenna can provide
useful diversity gain even at 7 MHz and some gain at 3.6 MHz.

Jack K8ZOA

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[David Cutter]

It did cross my mind that if you just want to verify, say the elevation
angle and intensity of the main lobe, you could launch a balloon some
wavelengths away carrying a field strength indicator.  With everything else
non-metal, the meter could send data via fibre to the ground and make a
measurement at every metre/foot.  Then move the balloon to another spot on
your plan and repeat.  Repeat for each antenna.  As long as it doesn't rain,
I would think that would give consistant results and it would take into
account the ground around the station.

David
G3UNA

>
>
> Geoff,
>
> To do good antenna testing you need an anechoic chamber or better
> operate in deep space. My primary interest in radio are the microwave
> bands. I do set up antenna ranges for 10 GHz (3cm), In doing so, I find
> the wind blowing tree leaves will have a non trivial affect on the
> observed signal even though the trees are "out of the way." I find that
> I need to place the reference antenna and the antenna under test in
> exactly the same place (x y and z position) to get true A/B comparisons.
> This is demonstrated by taking two known antenna and see how repeatable
> the observations can be.
>
> Scaling these realizations up to HF antenna causes me to throw up my
> hands. Yes we do the best we can testing antenna, but we have to be
> realistic about the data we collect. When anyone makes claims about
> their antenna I listen carefully for the details on how the antenna was
> tested.
>
> Maybe we should ask ARRL to fund a synchronous satellite over the US
> with a beacon on each of the HF bands so that we can point our antenna
> up and make more valid comparisons.
>
> Don, N0YE
>
>

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html