8405a working and measuring resonance?
 

Frank, this is consistent with what I am observing. The claim is for a given
frequency when a 0j load (in the range of 10 to 40 Ohms) and an unknown load
(assumed to be in the range of 10 to 40 Ohms) show the same phase shift then the
unknown is 0j also. It is not necessary that the loads be the same value, only
within the range. (I will have to experiment with loads over 60.)

So far nothing has surfaced to contradict this.

Thanks - Dan

Your method will exhibit a constant reflection coefficient circle (and
VSWR), with the angle varying from 180 degrees through zero degrees and then
back, through negative angles, to 180 degrees.

Frank

8405a working and measuring resonance?
 

"dansawyeror" wrote in message
...
Frank, this is consistent with what I am observing. The claim is for a
given frequency when a 0j load (in the range of 10 to 40 Ohms) and an
unknown load (assumed to be in the range of 10 to 40 Ohms) show the same
phase shift then the unknown is 0j also. It is not necessary that the
loads be the same value, only within the range. (I will have to experiment
with loads over 60.)

So far nothing has surfaced to contradict this.

Thanks - Dan

Dan, essentially your assumptions are correct providing you have no error
terms such as poor directivity.

Frank

8405a working and measuring resonance?
 

Thanks. I have two coupler systems, one with very good directivity and one with
only moderate directivity. The antenna is resonant at about 141.6 MHz. There is
a difference of about a couple of hundred kc between the systems.

Now on the the "R" measurements. Both systems indicate an antenna resistance of
over 30 Ohms. This antenna has been modeled with three different models. They
all indicate a Rr in the range of 5 Ohms. The antenna is grounded by a 3 foot by
5 foot screen mesh. While this is not infinite, it should provide a good ground.
Do you have a suggestion of where to look for the discrepancy?

Dan

Frank wrote:
"dansawyeror" wrote in message
...

Frank, this is consistent with what I am observing. The claim is for a
given frequency when a 0j load (in the range of 10 to 40 Ohms) and an
unknown load (assumed to be in the range of 10 to 40 Ohms) show the same
phase shift then the unknown is 0j also. It is not necessary that the
loads be the same value, only within the range. (I will have to experiment
with loads over 60.)

So far nothing has surfaced to contradict this.

Thanks - Dan


Dan, essentially your assumptions are correct providing you have no error
terms such as poor directivity.

Frank

8405a working and measuring resonance?
 

"dansawyeror" wrote in message
...
Thanks. I have two coupler systems, one with very good directivity and one
with only moderate directivity. The antenna is resonant at about 141.6
MHz. There is a difference of about a couple of hundred kc between the
systems.

Now on the the "R" measurements. Both systems indicate an antenna
resistance of over 30 Ohms. This antenna has been modeled with three
different models. They all indicate a Rr in the range of 5 Ohms. The
antenna is grounded by a 3 foot by 5 foot screen mesh. While this is not
infinite, it should provide a good ground. Do you have a suggestion of
where to look for the discrepancy?

Dan

Dan, the frequency difference is most likely caused by the poor directivity
of one of your couplers, so would not worry about it. Just take the reading
with the better coupler. The forward power coupled into the reverse port
will cause a phase error.

As for your modelling: what program are you using, and how do you describe
the ground screen? I ran a quick model above a perfect ground -- using
NEC2. According to my program the input impedance is 36 ohms at resonance.
The radiation efficiency, with a copper conductor, is 99.6%, so the
radiation resistance is obviously very close to 36 ohms. I can run a model
with a ground screen, but would like to replicate your structure as close as
possible. The actual height of my model monopole is 0.505 m (19.88"), and
is resonant at 141.8 MHz.

Frank

8405a working and measuring resonance?
 

Frank,

The antenna I am trying to model is a center 'loaded vertical'. It is a 4 inch
base, 5 turns at 40 percent spacing on a .8 diameter inch form and a 4 inch tip.
The material is Num 10 solid copper. I adjust the frequency by stretching or
compressing the coil. Currently it is resonant at about 141.7 Mhz. The 8405a
shows a phase shift of 1 degree per 30 kc change in frequency.

I have used both the vertload model and the EZNEC model. Both predict an antenna
R of about 5 Ohms. The 25 Ohm load shows a 12 db power difference between
forward and reverse. The antenna shows a 10 db power difference between forward
and reverse.

Thanks - Dan

Frank wrote:
"dansawyeror" wrote in message
...

Thanks. I have two coupler systems, one with very good directivity and one
with only moderate directivity. The antenna is resonant at about 141.6
MHz. There is a difference of about a couple of hundred kc between the
systems.

Now on the the "R" measurements. Both systems indicate an antenna
resistance of over 30 Ohms. This antenna has been modeled with three
different models. They all indicate a Rr in the range of 5 Ohms. The
antenna is grounded by a 3 foot by 5 foot screen mesh. While this is not
infinite, it should provide a good ground. Do you have a suggestion of
where to look for the discrepancy?

Dan


Dan, the frequency difference is most likely caused by the poor directivity
of one of your couplers, so would not worry about it. Just take the reading
with the better coupler. The forward power coupled into the reverse port
will cause a phase error.

As for your modelling: what program are you using, and how do you describe
the ground screen? I ran a quick model above a perfect ground -- using
NEC2. According to my program the input impedance is 36 ohms at resonance.
The radiation efficiency, with a copper conductor, is 99.6%, so the
radiation resistance is obviously very close to 36 ohms. I can run a model
with a ground screen, but would like to replicate your structure as close as
possible. The actual height of my model monopole is 0.505 m (19.88"), and
is resonant at 141.8 MHz.

Frank

8405a working and measuring resonance?
 

"dansawyeror" wrote in message
...
Frank,

The antenna I am trying to model is a center 'loaded vertical'. It is a 4
inch base, 5 turns at 40 percent spacing on a .8 diameter inch form and a
4 inch tip. The material is Num 10 solid copper. I adjust the frequency by
stretching or compressing the coil. Currently it is resonant at about
141.7 Mhz. The 8405a shows a phase shift of 1 degree per 30 kc change in
frequency.

I have used both the vertload model and the EZNEC model. Both predict an
antenna R of about 5 Ohms. The 25 Ohm load shows a 12 db power difference
between forward and reverse. The antenna shows a 10 db power difference
between forward and reverse.

Thanks - Dan

Dan, I have modelled a 5 turn inductor, 0.8" diameter, varying in length
from 0.8" to 1.6". The inductance values are 380 - 490 nH. An,
approximately 9" long monopole, with a 5 turn helix appears to be resonant
at about 190 MHz, with a highly reactive 6 ohm input impedance at 141 MHz.
Using a lumped element simulation the required load inductance, for 141 MHz,
is about 600 nH.

The only way to resolve these discrepancies is to do a standard single port
network analyzer calibration and measure the actual input impedance of the
antenna.

Frank

8405a working and measuring resonance?
 

Frank,

I will re-measure the coil dimensions. The recall the coil measured 600nH. That
is the value I used when I modeled this antenna using EZNec. It showed resonance
at about 145 MHz and 12 Ohms. (That was using an average real ground.)

If I assume the antenna measurements are correct then is it the ground that
accounts for the difference between 6 Ohms and the modeled 12 Ohms?

Now I am on to model and measure a center loaded dipole.

Dan

Frank wrote:
"dansawyeror" wrote in message
...

Frank,

The antenna I am trying to model is a center 'loaded vertical'. It is a 4
inch base, 5 turns at 40 percent spacing on a .8 diameter inch form and a
4 inch tip. The material is Num 10 solid copper. I adjust the frequency by
stretching or compressing the coil. Currently it is resonant at about
141.7 Mhz. The 8405a shows a phase shift of 1 degree per 30 kc change in
frequency.

I have used both the vertload model and the EZNEC model. Both predict an
antenna R of about 5 Ohms. The 25 Ohm load shows a 12 db power difference
between forward and reverse. The antenna shows a 10 db power difference
between forward and reverse.

Thanks - Dan


Dan, I have modelled a 5 turn inductor, 0.8" diameter, varying in length
from 0.8" to 1.6". The inductance values are 380 - 490 nH. An,
approximately 9" long monopole, with a 5 turn helix appears to be resonant
at about 190 MHz, with a highly reactive 6 ohm input impedance at 141 MHz.
Using a lumped element simulation the required load inductance, for 141 MHz,
is about 600 nH.

The only way to resolve these discrepancies is to do a standard single port
network analyzer calibration and measure the actual input impedance of the
antenna.

Frank

8405a working and measuring resonance?
 

Dan,

Be interested to see what the exact dimensions of the coil are. Anyway, it
seems we have some agreement on the 600 nH value, although my physical NEC
helix models do not agree based on my estimate of your coil dimensions. I
understand that EZNec uses a "Minninec" ground, which allows antenna contact
with a perfect ground, but uses actual ground parameters to analyze the
reflections. I am not sure about this, but would assume from the point of
view of the input impedance, that the ground would be considered perfect;
and therefore lossless. I also noticed I had some borderline NEC warnings
when attempting to construct a coil with #10 AWG, so sometimes had to resort
to a much thinner conductor.

My models showed about 17 ohms at resonance when connected to a perfect
ground. The only time I observed impedances as low as 6 ohms was far from
resonance when the antenna was highly capacitive. Ideally I should
construct a ground screen, but for the time being will consider a perfect
ground.

A free space dipole might be easier to model, but I am curious to understand
why there are discrepancies in the monopole modeling.

Frank

PS, be interested in any comments on my NEC code:

CM Loaded 2 m monopole
CE
GW 1 15 0.4 0 5.6 0.4 0 1.6 0.025
GH 2 50 .32 1.6 .4 .4 .4 .4 0.025
GM 100 0 0 0 0 0 0 4 000.051
GW 3 15 0.4 0 4 0.4 0 0 0.025
GS 0 0 0.025400
GE 1
GN 1
EX 0 3 15 00 1 0
LD 5 101 1 15 5.8001E7
LD 5 102 1 50 5.8001E7
LD 5 3 1 15 5.8001E7
FR 0 41 0 0 135 2
RP 0 181 1 1000 -90 90 1.00000 1.00000
EN
"dansawyeror" wrote in message
...
Frank,

I will re-measure the coil dimensions. The recall the coil measured 600nH.
That is the value I used when I modeled this antenna using EZNec. It
showed resonance at about 145 MHz and 12 Ohms. (That was using an average
real ground.)

If I assume the antenna measurements are correct then is it the ground
that accounts for the difference between 6 Ohms and the modeled 12 Ohms?

Now I am on to model and measure a center loaded dipole.

Dan

Frank wrote:
"dansawyeror" wrote in message
...

Frank,

The antenna I am trying to model is a center 'loaded vertical'. It is a 4
inch base, 5 turns at 40 percent spacing on a .8 diameter inch form and a
4 inch tip. The material is Num 10 solid copper. I adjust the frequency
by stretching or compressing the coil. Currently it is resonant at about
141.7 Mhz. The 8405a shows a phase shift of 1 degree per 30 kc change in
frequency.

I have used both the vertload model and the EZNEC model. Both predict an
antenna R of about 5 Ohms. The 25 Ohm load shows a 12 db power difference
between forward and reverse. The antenna shows a 10 db power difference
between forward and reverse.

Thanks - Dan


Dan, I have modelled a 5 turn inductor, 0.8" diameter, varying in length
from 0.8" to 1.6". The inductance values are 380 - 490 nH. An,
approximately 9" long monopole, with a 5 turn helix appears to be
resonant at about 190 MHz, with a highly reactive 6 ohm input impedance
at 141 MHz. Using a lumped element simulation the required load
inductance, for 141 MHz, is about 600 nH.

The only way to resolve these discrepancies is to do a standard single
port network analyzer calibration and measure the actual input impedance
of the antenna.

Frank

8405a working and measuring resonance?
 

Frank,

I will take the coil measurements.

Into which model programs can this description be loaded? Is there a 'howto'. I
looked on how to dump my model from EZNec and could not find it.

Dan


Frank wrote:
Dan,

Be interested to see what the exact dimensions of the coil are. Anyway, it
seems we have some agreement on the 600 nH value, although my physical NEC
helix models do not agree based on my estimate of your coil dimensions. I
understand that EZNec uses a "Minninec" ground, which allows antenna contact
with a perfect ground, but uses actual ground parameters to analyze the
reflections. I am not sure about this, but would assume from the point of
view of the input impedance, that the ground would be considered perfect;
and therefore lossless. I also noticed I had some borderline NEC warnings
when attempting to construct a coil with #10 AWG, so sometimes had to resort
to a much thinner conductor.

My models showed about 17 ohms at resonance when connected to a perfect
ground. The only time I observed impedances as low as 6 ohms was far from
resonance when the antenna was highly capacitive. Ideally I should
construct a ground screen, but for the time being will consider a perfect
ground.

A free space dipole might be easier to model, but I am curious to understand
why there are discrepancies in the monopole modeling.

Frank

PS, be interested in any comments on my NEC code:

CM Loaded 2 m monopole
CE
GW 1 15 0.4 0 5.6 0.4 0 1.6 0.025
GH 2 50 .32 1.6 .4 .4 .4 .4 0.025
GM 100 0 0 0 0 0 0 4 000.051
GW 3 15 0.4 0 4 0.4 0 0 0.025
GS 0 0 0.025400
GE 1
GN 1
EX 0 3 15 00 1 0
LD 5 101 1 15 5.8001E7
LD 5 102 1 50 5.8001E7
LD 5 3 1 15 5.8001E7
FR 0 41 0 0 135 2
RP 0 181 1 1000 -90 90 1.00000 1.00000
EN
"dansawyeror" wrote in message
...

Frank,

I will re-measure the coil dimensions. The recall the coil measured 600nH.
That is the value I used when I modeled this antenna using EZNec. It
showed resonance at about 145 MHz and 12 Ohms. (That was using an average
real ground.)

If I assume the antenna measurements are correct then is it the ground
that accounts for the difference between 6 Ohms and the modeled 12 Ohms?

Now I am on to model and measure a center loaded dipole.

Dan

Frank wrote:

"dansawyeror" wrote in message
...


Frank,

The antenna I am trying to model is a center 'loaded vertical'. It is a 4
inch base, 5 turns at 40 percent spacing on a .8 diameter inch form and a
4 inch tip. The material is Num 10 solid copper. I adjust the frequency
by stretching or compressing the coil. Currently it is resonant at about
141.7 Mhz. The 8405a shows a phase shift of 1 degree per 30 kc change in
frequency.

I have used both the vertload model and the EZNEC model. Both predict an
antenna R of about 5 Ohms. The 25 Ohm load shows a 12 db power difference
between forward and reverse. The antenna shows a 10 db power difference
between forward and reverse.

Thanks - Dan


Dan, I have modelled a 5 turn inductor, 0.8" diameter, varying in length
from 0.8" to 1.6". The inductance values are 380 - 490 nH. An,
approximately 9" long monopole, with a 5 turn helix appears to be
resonant at about 190 MHz, with a highly reactive 6 ohm input impedance
at 141 MHz. Using a lumped element simulation the required load
inductance, for 141 MHz, is about 600 nH.

The only way to resolve these discrepancies is to do a standard single
port network analyzer calibration and measure the actual input impedance
of the antenna.

Frank

8405a working and measuring resonance?
 

Frank,

I will take the coil measurements.

Into which model programs can this description be loaded? Is there a 'howto'. I
looked on how to dump my model from EZNec and could not find it.

Dan


Frank wrote:
Dan,

Be interested to see what the exact dimensions of the coil are. Anyway, it
seems we have some agreement on the 600 nH value, although my physical NEC
helix models do not agree based on my estimate of your coil dimensions. I
understand that EZNec uses a "Minninec" ground, which allows antenna contact
with a perfect ground, but uses actual ground parameters to analyze the
reflections. I am not sure about this, but would assume from the point of
view of the input impedance, that the ground would be considered perfect;
and therefore lossless. I also noticed I had some borderline NEC warnings
when attempting to construct a coil with #10 AWG, so sometimes had to resort
to a much thinner conductor.

My models showed about 17 ohms at resonance when connected to a perfect
ground. The only time I observed impedances as low as 6 ohms was far from
resonance when the antenna was highly capacitive. Ideally I should
construct a ground screen, but for the time being will consider a perfect
ground.

A free space dipole might be easier to model, but I am curious to understand
why there are discrepancies in the monopole modeling.

Frank

PS, be interested in any comments on my NEC code:

CM Loaded 2 m monopole
CE
GW 1 15 0.4 0 5.6 0.4 0 1.6 0.025
GH 2 50 .32 1.6 .4 .4 .4 .4 0.025
GM 100 0 0 0 0 0 0 4 000.051
GW 3 15 0.4 0 4 0.4 0 0 0.025
GS 0 0 0.025400
GE 1
GN 1
EX 0 3 15 00 1 0
LD 5 101 1 15 5.8001E7
LD 5 102 1 50 5.8001E7
LD 5 3 1 15 5.8001E7
FR 0 41 0 0 135 2
RP 0 181 1 1000 -90 90 1.00000 1.00000
EN
"dansawyeror" wrote in message
...

Frank,

I will re-measure the coil dimensions. The recall the coil measured 600nH.
That is the value I used when I modeled this antenna using EZNec. It
showed resonance at about 145 MHz and 12 Ohms. (That was using an average
real ground.)

If I assume the antenna measurements are correct then is it the ground
that accounts for the difference between 6 Ohms and the modeled 12 Ohms?

Now I am on to model and measure a center loaded dipole.

Dan

Frank wrote:

"dansawyeror" wrote in message
...


Frank,

The antenna I am trying to model is a center 'loaded vertical'. It is a 4
inch base, 5 turns at 40 percent spacing on a .8 diameter inch form and a
4 inch tip. The material is Num 10 solid copper. I adjust the frequency
by stretching or compressing the coil. Currently it is resonant at about
141.7 Mhz. The 8405a shows a phase shift of 1 degree per 30 kc change in
frequency.

I have used both the vertload model and the EZNEC model. Both predict an
antenna R of about 5 Ohms. The 25 Ohm load shows a 12 db power difference
between forward and reverse. The antenna shows a 10 db power difference
between forward and reverse.

Thanks - Dan


Dan, I have modelled a 5 turn inductor, 0.8" diameter, varying in length
from 0.8" to 1.6". The inductance values are 380 - 490 nH. An,
approximately 9" long monopole, with a 5 turn helix appears to be
resonant at about 190 MHz, with a highly reactive 6 ohm input impedance
at 141 MHz. Using a lumped element simulation the required load
inductance, for 141 MHz, is about 600 nH.

The only way to resolve these discrepancies is to do a standard single
port network analyzer calibration and measure the actual input impedance
of the antenna.

Frank