John,
I had breakfast with some friends this morning. One of them has a PHD in EE,
specializing in antenna design. He thinks the antenna should work, but
suggested changing the spacing between the upper and lower conductors. He
also told me to look for multiple resonances. I just now tried that, with
the spacing essentially 0. What I got was a new resonance at 165 MHz, with
Z=18 + j0. . Interestingly, I now seem to have hit parallel resonance, like
you did, and R goes down either side of 165. In fact, accross 100 -150 MHz
the Z is 0 -jX. When I was messing around with spacing, at one time I got
202 + j0 at 129 MHz; but that is way off calculated frequency. I probably
should not be doing this inside, as it is not entirely repeatable.

I also tried a longer piece of coax (RG8X, Vp=.75) that gave me a 5 foot
folded dipole. That should be resonant at about 93 MHz. I clearly got
multiple resonances:

F R X
92.5 3 j3
100 11 j44
110 80 j181
120 4 j30
127 4 j1
160 53 j14 SWR=1.3
170 11 j0

The feedline was also 5 feet, since I did not know what a wavelengt was
going to be. This looks like it wants to be a 160 MHz antenna, instead of
93. Lastly, I tried the 5 foot antennawith a 2 inch feedline.Resonance was
at 156 MHz. If I get a chance, I will try it outdoors tomorrow.

Tam/WB2TT

On Mon, 4 Oct 2004 20:58:07 -0500, "John Smith"
wrote:

|
|"Wes Stewart" wrote in message
.. .
| On Mon, 4 Oct 2004 16:45:31 -0500, "John Smith"
| wrote:
|
|Freq (MHz) Impedance (Ohms)
|
|410 46.4 + 6.0i (50 Ohm terminator)
|
|380 9.7 - 12.5
|390 3.5 - 5.7
|400 5.1 + 1.3i
|410 5.1 + 6.5i
|415 4.0 + 10.0i
|425 2.5 + 15.7i
|
|
| | 3) How well is your signal source terminated, in other words do you
| | know its source match?
| |
| |
| |I only know that the signal source is an HP 3200B. It directly feeds the
| |Narda dual directional coupler through a few feet of RG58.
|
| If I remember that correctly the '3200 is nothing but a p-p oscillator
| and a waveguide-below-cutoff probe. If your VVM reference probe
| readings are changing much between frequencies and/or
| calibration/measurement, try a 6 or 10 dB pad right on the generator
| output and see what happens.
|
| When you're calibrating using a short, the source Z has really got to
| be nailed down.
|
| Wes
|
|
|Okay. I repeated the test using an HP 355C attenuator set for 10 dB and at
|400 MHz got 4 + 3i on the antenna. I also checked my 50 Ohm network
|terminator with this setup and it measured 44 + 4i. The data are different,
|but they're not an order of magnitude different, at least.
|
|So, although my measurements aren't repeatable, they are sloppily
|consistent. That is, although I can't say exactly what the antenna impedance
|is with confidence, I am beginning to believe that it really is very low in
|impedance. Am I drawing an erroneous conclusion too early? I can wait a
|little longer to draw an erroneous conclusion.

First of all, neglecting the feed method, the antenna is a simple
folded dipole. In free space, or an approximation thereof, it should
have a feedpoint Z of about 300 ohm. (See the ARRL Antenna book for a
description of why this is so under "Special Antenna Types", p.2-32 in
the 17th edition) In the presence of other (non-resonant) objects, it
may differ from this but not a whole lot.

In theory, the "natural balun" doesn't change the impedance of the
feedpoint whatsoever. By "feedpoint" I mean the gap between the ends
of the folded element, not the "tee" connection opposite.

At the outside of the tee connection, the voltage is zero thus this
point can be grounded, connected to the boom in a Yagi, etc. without
upsetting anything. Likewise the coax feeder can be introduced here
and run through one side of the element without upsetting anything
either.

But, a nominal 300 ohm load is terminating a 50 ohm line, so the usual
transforming effects are in play. The input Z of an arbitrary length
line is---well, arbitrary. If the line is many wavelengths long, then
when the frequency is changed, the long lines effect kicks in and the
input Z is going to vary rapidly with respect to frequency.

Second. I believe that you need to determine the parameters of your
directional coupler. As Richard pointed out, your B1/A1 numbers are
pretty unstable.

So here's what I recommend. First verify that the "A" and "B" probes
read the same thing when connected to the same source. Then put your
pad right at the input connector of the coupler. Terminate the
reverse port and connect your VVM "A" probe to the ouput connector and
the "B" probe to the forward port.

The ratio reading is the forward coupling factor of the directional
coupler. Vary the frequency and see how this changes and note some
values. Move the "B" probe to the reverse port and terminate the
forward port. Note the readings at the same frequency.

Reverse the input and output ports and repeat the measurements.
Ideally, the data sets will track closely. If they don't then you
have a problem. Serious differences might indicate damage to the
internal terminations. This assumes that this is a true dual coupler
and not single line coupler with the termination applied to the unused
port externally.

If the numbers are consistant, then you can determine the directivity
by computing the ratio between the two readings on a given port when
the coupler is reversed.

I'm going to stop here and assume you understand the consquences of
poor directivity on measurement accuracy. If you don't then I can
expound further later.

Wes

"Tam/WB2TT" wrote in message
...
John,
I had breakfast with some friends this morning. One of them has a PHD in
EE, specializing in antenna design. He thinks the antenna should work, but
suggested changing the spacing between the upper and lower conductors. He
also told me to look for multiple resonances. I just now tried that, with
the spacing essentially 0. What I got was a new resonance at 165 MHz, with
Z=18 + j0. . Interestingly, I now seem to have hit parallel resonance,
like you did, and R goes down either side of 165. In fact, accross
100 -150 MHz the Z is 0 -jX. When I was messing around with spacing, at
one time I got 202 + j0 at 129 MHz; but that is way off calculated
frequency. I probably should not be doing this inside, as it is not
entirely repeatable.

I also tried a longer piece of coax (RG8X, Vp=.75) that gave me a 5 foot
folded dipole. That should be resonant at about 93 MHz. I clearly got
multiple resonances:

F R X
92.5 3 j3
100 11 j44
110 80 j181
120 4 j30
127 4 j1
160 53 j14 SWR=1.3
170 11 j0

The feedline was also 5 feet, since I did not know what a wavelengt was
going to be. This looks like it wants to be a 160 MHz antenna, instead of
93. Lastly, I tried the 5 foot antennawith a 2 inch feedline.Resonance was
at 156 MHz. If I get a chance, I will try it outdoors tomorrow.

Tam/WB2TT


Thanks for the hard work, Tam. I'm not sure I know what to make of all this,
but it appears that the antenna is not what I thought it would be.

Thanks again.

John

"Wes Stewart" wrote in message
...
On Mon, 4 Oct 2004 20:58:07 -0500, "John Smith"
wrote:

|
|"Wes Stewart" wrote in message
.. .
| On Mon, 4 Oct 2004 16:45:31 -0500, "John Smith"
| wrote:
|
|Freq (MHz) Impedance (Ohms)
|
|410 46.4 + 6.0i (50 Ohm terminator)
|
|380 9.7 - 12.5
|390 3.5 - 5.7
|400 5.1 + 1.3i
|410 5.1 + 6.5i
|415 4.0 + 10.0i
|425 2.5 + 15.7i
|
|
| | 3) How well is your signal source terminated, in other words do you
| | know its source match?
| |
| |
| |I only know that the signal source is an HP 3200B. It directly feeds
the
| |Narda dual directional coupler through a few feet of RG58.
|
| If I remember that correctly the '3200 is nothing but a p-p oscillator
| and a waveguide-below-cutoff probe. If your VVM reference probe
| readings are changing much between frequencies and/or
| calibration/measurement, try a 6 or 10 dB pad right on the generator
| output and see what happens.
|
| When you're calibrating using a short, the source Z has really got to
| be nailed down.
|
| Wes
|
|
|Okay. I repeated the test using an HP 355C attenuator set for 10 dB and
at
|400 MHz got 4 + 3i on the antenna. I also checked my 50 Ohm network
|terminator with this setup and it measured 44 + 4i. The data are
different,
|but they're not an order of magnitude different, at least.
|
|So, although my measurements aren't repeatable, they are sloppily
|consistent. That is, although I can't say exactly what the antenna
impedance
|is with confidence, I am beginning to believe that it really is very low
in
|impedance. Am I drawing an erroneous conclusion too early? I can wait a
|little longer to draw an erroneous conclusion.

First of all, neglecting the feed method, the antenna is a simple
folded dipole. In free space, or an approximation thereof, it should
have a feedpoint Z of about 300 ohm. (See the ARRL Antenna book for a
description of why this is so under "Special Antenna Types", p.2-32 in
the 17th edition) In the presence of other (non-resonant) objects, it
may differ from this but not a whole lot.

In theory, the "natural balun" doesn't change the impedance of the
feedpoint whatsoever. By "feedpoint" I mean the gap between the ends
of the folded element, not the "tee" connection opposite.

At the outside of the tee connection, the voltage is zero thus this
point can be grounded, connected to the boom in a Yagi, etc. without
upsetting anything. Likewise the coax feeder can be introduced here
and run through one side of the element without upsetting anything
either.

But, a nominal 300 ohm load is terminating a 50 ohm line, so the usual
transforming effects are in play. The input Z of an arbitrary length
line is---well, arbitrary. If the line is many wavelengths long, then
when the frequency is changed, the long lines effect kicks in and the
input Z is going to vary rapidly with respect to frequency.

Second. I believe that you need to determine the parameters of your
directional coupler. As Richard pointed out, your B1/A1 numbers are
pretty unstable.

So here's what I recommend. First verify that the "A" and "B" probes
read the same thing when connected to the same source. Then put your
pad right at the input connector of the coupler. Terminate the
reverse port and connect your VVM "A" probe to the ouput connector and
the "B" probe to the forward port.

The ratio reading is the forward coupling factor of the directional
coupler. Vary the frequency and see how this changes and note some
values. Move the "B" probe to the reverse port and terminate the
forward port. Note the readings at the same frequency.

Reverse the input and output ports and repeat the measurements.
Ideally, the data sets will track closely. If they don't then you
have a problem. Serious differences might indicate damage to the
internal terminations. This assumes that this is a true dual coupler
and not single line coupler with the termination applied to the unused
port externally.

If the numbers are consistant, then you can determine the directivity
by computing the ratio between the two readings on a given port when
the coupler is reversed.

I'm going to stop here and assume you understand the consquences of
poor directivity on measurement accuracy. If you don't then I can
expound further later.

Wes

Thanks, Wes. I'm going to have to stop the experiments for a few days, but
I'll try to get back to you.

John

"Wes Stewart" wrote in message
...
On Mon, 4 Oct 2004 20:58:07 -0500, "John Smith"
wrote:

Second. I believe that you need to determine the parameters of your
directional coupler. As Richard pointed out, your B1/A1 numbers are
pretty unstable.

So here's what I recommend. First verify that the "A" and "B" probes
read the same thing when connected to the same source.

I put the oscillator to the center of the HP Power Splitter. I then put a
probe tee on each of the splitter outputs followed by a 50 Ohm terminator.
The difference between the A and B channels was maybe a needle's width.
Reversing the splitter made no difference. Swapping the terminators made no
difference.


Then put your
pad right at the input connector of the coupler. Terminate the
reverse port and connect your VVM "A" probe to the ouput connector and
the "B" probe to the forward port.

The ratio reading is the forward coupling factor of the directional
coupler. Vary the frequency and see how this changes and note some
values. Move the "B" probe to the reverse port and terminate the
forward port. Note the readings at the same frequency.

From 350 MHz to 450 MHz the forward port coupling was -25.9 dB to -26.2 dB.

Reverse the input and output ports and repeat the measurements.
Ideally, the data sets will track closely.


I got the same here within about a tenth of a dB.


If they don't then you
have a problem. Serious differences might indicate damage to the
internal terminations. This assumes that this is a true dual coupler
and not single line coupler with the termination applied to the unused
port externally.

If the numbers are consistant, then you can determine the directivity
by computing the ratio between the two readings on a given port when
the coupler is reversed.

I'm going to stop here and assume you understand the consquences of
poor directivity on measurement accuracy. If you don't then I can
expound further later.

Wes


I guess it looks okay. Thanks, Wes.

John (KD5YI)

On Tue, 5 Oct 2004 12:26:06 -0500, "John Smith"
wrote:

You should get 1.0 @ 180° (the definition of a short).

Well, that's not possible when feeding a length of RG58 at 400 MHz, is it?
Remember, I said that there was about a 5 foot length of RG58 between the
directional coupler and the load. How can one get 1.0 reflected to the
coupler when the load is a short? That requires zero loss coax.

Hi John,

As Wes suggests, butt up the load against the directional coupler
output and eliminate this arbitrary loss of the 5 foot RG58. It
should also shift the readings too (you are simply walking around the
circle of constant SWR). One question that would be obviated in this
process (but I have to ask anyway) is WHERE was this short you
applied? At the output port of the coupler, or at the end of this 5
foot RG58? (Same question applies to the calibrated load).

The other measurements that you reported in response to Wes indicate
you have tracking ports (even if they are off by 4dB). As I said, it
seemed unlikely this would be a problem and it confirms the
out-of-octave specification.

73's
Richard Clark, KB7QHC

On Tue, 5 Oct 2004 15:59:32 -0500, "John Smith"
wrote:

|
|"Wes Stewart" wrote in message
.. .
| On Mon, 4 Oct 2004 20:58:07 -0500, "John Smith"
| wrote:
|
| Second. I believe that you need to determine the parameters of your
| directional coupler. As Richard pointed out, your B1/A1 numbers are
| pretty unstable.
|
| So here's what I recommend. First verify that the "A" and "B" probes
| read the same thing when connected to the same source.
|
|I put the oscillator to the center of the HP Power Splitter. I then put a
|probe tee on each of the splitter outputs followed by a 50 Ohm terminator.
|The difference between the A and B channels was maybe a needle's width.
|Reversing the splitter made no difference. Swapping the terminators made no
|difference.

Excellent.
|
|
| Then put your
| pad right at the input connector of the coupler. Terminate the
| reverse port and connect your VVM "A" probe to the ouput connector and
| the "B" probe to the forward port.
|
| The ratio reading is the forward coupling factor of the directional
| coupler. Vary the frequency and see how this changes and note some
| values. Move the "B" probe to the reverse port and terminate the
| forward port. Note the readings at the same frequency.
|
|From 350 MHz to 450 MHz the forward port coupling was -25.9 dB to -26.2 dB.

Okay. Not per nameplate, but now you know.
|
| Reverse the input and output ports and repeat the measurements.
| Ideally, the data sets will track closely.
|
|
|I got the same here within about a tenth of a dB.

Great.
|
|
| If they don't then you
| have a problem. Serious differences might indicate damage to the
| internal terminations. This assumes that this is a true dual coupler
| and not single line coupler with the termination applied to the unused
| port externally.
|
| If the numbers are consistant, then you can determine the directivity
| by computing the ratio between the two readings on a given port when
| the coupler is reversed.

You still need to do this.

|
| I'm going to stop here and assume you understand the consquences of
| poor directivity on measurement accuracy. If you don't then I can
| expound further later.
|
| Wes
|
|
|I guess it looks okay. Thanks, Wes.

You're welcome.

Wes

"Wes Stewart" wrote in message
...
On Tue, 5 Oct 2004 15:59:32 -0500, "John Smith"
wrote:

| If the numbers are consistant, then you can determine the directivity
| by computing the ratio between the two readings on a given port when
| the coupler is reversed.

You still need to do this.



Do what? They are both -26.2 dB. I don't understand.

John

"Richard Clark" wrote in message
news
On Tue, 5 Oct 2004 12:26:06 -0500, "John Smith"
wrote:

You should get 1.0 @ 180° (the definition of a short).

Well, that's not possible when feeding a length of RG58 at 400 MHz, is it?
Remember, I said that there was about a 5 foot length of RG58 between the
directional coupler and the load. How can one get 1.0 reflected to the
coupler when the load is a short? That requires zero loss coax.

Hi John,

As Wes suggests, butt up the load against the directional coupler
output and eliminate this arbitrary loss of the 5 foot RG58. It
should also shift the readings too (you are simply walking around the
circle of constant SWR). One question that would be obviated in this
process (but I have to ask anyway) is WHERE was this short you
applied? At the output port of the coupler, or at the end of this 5
foot RG58? (Same question applies to the calibrated load).


I used a 66 inch piece of RG58 between the directional coupler and the load.
It was at the load end of this piece of coax that I calibrated with a short
and made the load measurements.


The other measurements that you reported in response to Wes indicate
you have tracking ports (even if they are off by 4dB). As I said, it
seemed unlikely this would be a problem and it confirms the
out-of-octave specification.

73's
Richard Clark, KB7QHC

"Wes Stewart" wrote in message
...
On Tue, 5 Oct 2004 15:59:32 -0500, "John Smith"
wrote:


| If the numbers are consistant, then you can determine the directivity
| by computing the ratio between the two readings on a given port when
| the coupler is reversed.

You still need to do this.


Hi, Wes -

I read a little about directivity (a little was all I could find). Tell me
if I measured it correctly...

Forward directivity:
Normal setup, ie coupler in the usual direction (forward). Best 50 Ohm load
I could muster on the antenna (output) terminal of the coupler. Set channel
A for 0 dB. Channel B reads -29.2 dB.

Forward Open/Short characteristic:
Remove the 50 Ohm load. Set channel A for 0 dB. Channel B reads +.4 dB. Put
the HP calibrated short on the antenna terminal. Set channel A for 0 dB.
Channel B reads -1.2 dB.

Reverse directivity:
Reverse the coupler. The antenna connector now has the oscillator applied.
The BTS terminal has the 50 Ohm load. Set channel A for 0 dB. Channel B
reads -23 dB.

Reverse Open/Short characteristic:
Remove the 50 Ohm load. Set channel A for 0 dB. Channel B reads +1.1 dB. Put
the HP calibrated short on the BTS (now the output) terminal. Set channel A
for 0 dB. Channel B reads -.5 dB.

So, in the forward direction, the directivity is -29.2 dB, and in the
reverse direction, the directivity is -23 dB. Yes?

What do I do with these results? I don't know how to apply them even though
I read the HP paper.

Thanks.

John (KD5YI)

By the way, if you (or anyone else) need to contact me via email, you can
omit the kes in the address.