I have a 1/2 wave - end fed whip antenna that has been designed for
operation on 915 MHz.

A Signal Generator is attached to a VSWR detector. The reference port is
50 Ohms. The antenna is connected to Z2 port and the detector output to
an oscilloscope.

I have calibrated the output for 50 Ohms and noted the 2:1 VSWR voltage when
using a 2:1 precision ref at Z2 in place of the antenna.

I have a Spectrum analyzer with short wire as a pickup to observe
radiation from antenna (I image 1/2 the trouble is here ?)

When I sweep manually across say 800 MHz to 1GHz the VSWR gets worse
when the radiated power increases and visa versa.

When I then plug in a commercial antenna the VSWR does the same thing
EXCEPT when it is in the region it is resonant, the VSWR improves as the
level rises, at all other points the VSWR goes down as radiated power
increases.

Is there something silly going on here or is this what should happen.
This is the first time I have attempted to measure an antenna so would
appreciate any help.
I notice the peaks at a reasonably fixed period
(816,879,933,993,1.051G), can I learn something about the antenna from
how apart these peaks ?

Thanks in advance,

David wrote:
I have a 1/2 wave - end fed whip antenna that has been designed for
operation on 915 MHz.

A Signal Generator is attached to a VSWR detector. The reference port is
50 Ohms. The antenna is connected to Z2 port and the detector output to
an oscilloscope.

I have calibrated the output for 50 Ohms and noted the 2:1 VSWR voltage
when
using a 2:1 precision ref at Z2 in place of the antenna.

I have a Spectrum analyzer with short wire as a pickup to observe
radiation from antenna (I image 1/2 the trouble is here ?)

When I sweep manually across say 800 MHz to 1GHz the VSWR gets worse
when the radiated power increases and visa versa.

When I then plug in a commercial antenna the VSWR does the same thing
EXCEPT when it is in the region it is resonant, the VSWR improves as the
level rises, at all other points the VSWR goes down as radiated power
increases.

Is there something silly going on here or is this what should happen.
This is the first time I have attempted to measure an antenna so would
appreciate any help.
I notice the peaks at a reasonably fixed period
(816,879,933,993,1.051G), can I learn something about the antenna from
how apart these peaks ?

Thanks in advance,

If the wire is significantly less than a wavelength on the spectrum
analyzer it shouldn't cause effects that are _too_ weird, and is
probably better than hanging an unknown sorta-tuned antenna on the thing.

You mentioned the Magic Flame War Word when you mentioned VSWR. The
relationship between VSWR, amplifier performance and antenna performance
is complex. I'll get behind this here bunker and say that VSWR
measurements are useful things if you know the effects, and if you're
using components that are designed for 1:1 VSWR at some design impedance
that's what you should do. Beyond that a given amplifier may not have
it's peak power output at 1:1 VSWR, and a given antenna may not have
it's peak effectiveness at 1:1 VSWR, unless the amplifier and antenna
are specifically designed for that. Try to ignore the side threads that
this paragraph will generate -- those folks all voted for either Nader
or LaRouch in the last election.

Did you pay attention to the amplifier power output when you were doing
your VSWR sweep? Did you try moving the spectrum analyzer pickup
around? An amplifier that's designed for a 50 ohm output and run into a
different impedance may indeed generate more power, at least until the
finals overheat. It may be that the amplifier you're using happens to
generate more power into the off-tune load that the antenna presents -
or - that the antenna radiates more efficiently in the direction of your
pickup.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

David wrote:
I have a 1/2 wave - end fed whip antenna that has been designed for
operation on 915 MHz.

Last time I inquired about this antenna I was told it wasn't
an end-fed but instead was a center-fed sleeve antenna. So
which is it? (Anti)resonance on a bona fide end-fed 1/2WL
antenna is pretty hard to nail down but it doesn't much matter.
One just usually searches for a 50 ohm tap on a coil.
--
73, Cecil http://www.qsl.net/w5dxp


----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 100,000 Newsgroups
---= East/West-Coast Server Farms - Total Privacy via Encryption =---

A half-wave end-fed wire, fed against a ground plane or similar
counterpoise, will show a high impedance at (and near) resonance. If
the antenna is fed through some length of line from the "VSWR
detector," the impedance presented to the detector will be modified by
that length of line; a 1000 ohm antenna might reflect 3 ohms through
the line. But why would you expect a drop in indicated VSWR at
resonance? -- When you say "when the radiated power increases," are
you talking about increased field strength as measured by the spectrum
analyzer, with a constant output setting from the generator? How far
away is the spectrum analyzer antenna, in wavelengths, from the antenna
you are trying to measure?

If, as Cecil suspects, the antenna is really a center-fed half-wave,
what have you done to decouple the feedline from the antenna? If the
feedline comes coaxially out the end of the antenna, it may well be
that the feedline is very poorly decoupled.

Cheers,
Tom

I notice the peaks at a reasonably fixed period
(816,879,933,993,1.051G), can I learn something about the antenna from
how apart these peaks ?

David,

Judging from your comment above, it appears that the VSWR peaks are
happening at a very periodic spacing of about 60MHz. This would imply to me
that you're analyzing the antenna by looking into a feedline that is 1/4
wavelength long at 60 MHz. This would be about 49 inches times the velocity
factor of the coax. Thus I'd estimate that your feedline is in the mid 30"
range. Is this the case?

Joe
W3JDR



"David" wrote in message
...
I have a 1/2 wave - end fed whip antenna that has been designed for
operation on 915 MHz.

A Signal Generator is attached to a VSWR detector. The reference port is
50 Ohms. The antenna is connected to Z2 port and the detector output to
an oscilloscope.

I have calibrated the output for 50 Ohms and noted the 2:1 VSWR voltage
when
using a 2:1 precision ref at Z2 in place of the antenna.

I have a Spectrum analyzer with short wire as a pickup to observe
radiation from antenna (I image 1/2 the trouble is here ?)

When I sweep manually across say 800 MHz to 1GHz the VSWR gets worse
when the radiated power increases and visa versa.

When I then plug in a commercial antenna the VSWR does the same thing
EXCEPT when it is in the region it is resonant, the VSWR improves as the
level rises, at all other points the VSWR goes down as radiated power
increases.

Is there something silly going on here or is this what should happen.
This is the first time I have attempted to measure an antenna so would
appreciate any help.
I notice the peaks at a reasonably fixed period
(816,879,933,993,1.051G), can I learn something about the antenna from
how apart these peaks ?

Thanks in advance,

Sorry about my description of the antenna. I still have not yet come to
grips with the terms of reference. I called it end fed because the coax
enters at the bottom but I have been advised it is actually a centre fed
dipole.

The decoupling is via a 1/4 wave sleeve that provides high impedance for
RF returning along outer coax and also as the second 1/2 of the dipole.

The level I noted is the signal strength seen on a spectrum analyzer
(80mm wire at end of 1m cable) - this may be the 30" section that is
producing the 60MHz dips. I also noticed the wavelength of this pickup
wire is close to 1/4 wave at frequency of interest.

Would I be better to put a directional coupler in line with the Signal
Generator output and sniff off to spec analyzer and look at dips in
signal strength as indication of radiation occurring ? (I would like to
see what the radiated power is doing at the same time as the VSWR is all
over the place and hard to see what is happening when looking at that
only). If I can do away with the wire pickup on the spectrum analyzer I
can at least remove the effects of this being super imposed on the results.

The commercial antenna I am comparing against does have a peak radiated
power at min. VSWR at frequency of interest and my antenna does not.

The style of antenna is the same (1/4 wave shunt sleeve and 1/4 wave
radiating element).

Thanks for any more help.

Regards

David



K7ITM wrote:
A half-wave end-fed wire, fed against a ground plane or similar
counterpoise, will show a high impedance at (and near) resonance. If
the antenna is fed through some length of line from the "VSWR
detector," the impedance presented to the detector will be modified by
that length of line; a 1000 ohm antenna might reflect 3 ohms through
the line. But why would you expect a drop in indicated VSWR at
resonance? -- When you say "when the radiated power increases," are
you talking about increased field strength as measured by the spectrum
analyzer, with a constant output setting from the generator? How far
away is the spectrum analyzer antenna, in wavelengths, from the antenna
you are trying to measure?

If, as Cecil suspects, the antenna is really a center-fed half-wave,
what have you done to decouple the feedline from the antenna? If the
feedline comes coaxially out the end of the antenna, it may well be
that the feedline is very poorly decoupled.

Cheers,
Tom

"Cecil Moore" wrote in message
...
David wrote:
I have a 1/2 wave - end fed whip antenna that has been designed for
operation on 915 MHz.

Last time I inquired about this antenna I was told it wasn't
an end-fed but instead was a center-fed sleeve antenna. So
which is it? (Anti)resonance on a bona fide end-fed 1/2WL
antenna is pretty hard to nail down but it doesn't much matter.
One just usually searches for a 50 ohm tap on a coil.
--
73, Cecil http://www.qsl.net/w5dxp

Cecil

I have been in contact with David. I gave him sketches for building a
centerfed 1/2 wave dipole with the coax feed line running inside one half of
the dipole. I am sure he has constructed an antenna like that, center fed
half wave dipole.

Jerry

Hi David,

You wrote, "The decoupling is via a 1/4 wave sleeve that provides high
impedance for
RF returning along outer coax and also as the second 1/2 of the
dipole. "

EXACTLY how is this built? The details of construction make a BIG
difference in performance! (There's a lot of BAD info about it out
there...)

It's not a bad idea to ALSO put some additional decoupling further down
the feedline.

If your spectrum analyzer/field strength meter is far enough away from
the antenna you are testing, then it should provide a reasonable
indication of relative antenna radiation performance. The SWR
indication, if properly calibrated and given that you are apparently
exciting the antenna with a source whose output impedance matches your
feedline, should also be a good indication of power actually absorbed
by the antenna. That is, lowest SWR represents maximum power absorbed
by the antenna. Presumably that power is being radiated as RF, mostly,
and not dissipated as heat. But where the RF radiation goes depends on
the pattern of currents excited on the conductors that compose the
antenna, and nearby conductors as well (such as the feedline). What
you probably want is standard resonant half-wave dipole currents on
your vertical dipole, and no (very little) antenna current on the
feedline and on support structures. By the way, whether the antenna is
resonant or not is of little real importance, so long as you can
efficiently feed power to it and the antenna currents are in the right
places and not the wrong places. But it happens that with your
antenna, if things are working properly (properly decoupled feedline,
etc), you probably will see lowest SWR at half-wave resonance. If you
have no other matching going on, the lowest SWR will probably be about
1.5:1 with 50 ohm feedline. You could add parts to get a better match
if you wished.

And as you can probably tell from all that, I'm suspecting that your
decoupling sleeve, with associated dielectrics in that area, probably
isn't doing a very good job...

Also...Joe noted that your coax feedline may well be a length that
accounts for the SWR peaks and valleys. (I think it may be about twice
as long as Joe wrote...but same idea.) Do you see the peaks and
valleys when you terminate the line in the precision 2:1 load? If you
do NOT, then it's a further indication that the feedline has antenna
currents on it, because the flat 2:1 is an indication that your
transmission line is matched to the calibration impedance of the SWR
bridge, and if that's the case, the SWR bridge should be giving at
reasonably accurate estimate of the actual line SWR. If you DO see the
SWR ripples vs frequency with just the precision load, either the load
isn't "flat" or the line is not the same impedance as the SWR bridge is
calibrated to, and the differing impedances is by far the most probable
explanation if the line length is right.

Cheers,
Tom

Tom,

Thanks for the information. The inner coax is the smaller RG174 coax.
The sleeve is made of earth braid pulled from RG58 cable. The dielectric
between the sleeve and inner cable is therefore the outer sheath of the
RG174 cable (Not sure what this is, the RG174 I have is Teflon inner
dielectric and stranded conductor. The utter sheath is a very strong
heat resistant material - I therefore have no ideal of the dielectric
constant to calculate Vp for correct electrical length). If I use a
copper tube and strip off the sheath from the inner coax, I can
calculate correct length as it will have an air dielectric.

Do you know where everyone is getting the dielectric constants for
various materials ? I noted people using small metal tubes as sleeves
and quoting these magic numbers even for copper tubes of certain diameters.

From my discussion with Telonic, they say the Rho_Tector was designed
as an in-house tool for measuring inputs of amps and filters. They
suggest SWR meter would probably be best for antenna adjustments.
Do you happen to know where I might find details for a low power SWR
meter for 915 MHz ? I need one that will operate with only 20mW applied
power. The only SWR meter I have has min. FSD of 3W

Thanks

Regards

David

K7ITM wrote:
Hi David,

You wrote, "The decoupling is via a 1/4 wave sleeve that provides high
impedance for
RF returning along outer coax and also as the second 1/2 of the
dipole. "

EXACTLY how is this built? The details of construction make a BIG
difference in performance! (There's a lot of BAD info about it out
there...)

It's not a bad idea to ALSO put some additional decoupling further down
the feedline.

If your spectrum analyzer/field strength meter is far enough away from
the antenna you are testing, then it should provide a reasonable
indication of relative antenna radiation performance. The SWR
indication, if properly calibrated and given that you are apparently
exciting the antenna with a source whose output impedance matches your
feedline, should also be a good indication of power actually absorbed
by the antenna. That is, lowest SWR represents maximum power absorbed
by the antenna. Presumably that power is being radiated as RF, mostly,
and not dissipated as heat. But where the RF radiation goes depends on
the pattern of currents excited on the conductors that compose the
antenna, and nearby conductors as well (such as the feedline). What
you probably want is standard resonant half-wave dipole currents on
your vertical dipole, and no (very little) antenna current on the
feedline and on support structures. By the way, whether the antenna is
resonant or not is of little real importance, so long as you can
efficiently feed power to it and the antenna currents are in the right
places and not the wrong places. But it happens that with your
antenna, if things are working properly (properly decoupled feedline,
etc), you probably will see lowest SWR at half-wave resonance. If you
have no other matching going on, the lowest SWR will probably be about
1.5:1 with 50 ohm feedline. You could add parts to get a better match
if you wished.

And as you can probably tell from all that, I'm suspecting that your
decoupling sleeve, with associated dielectrics in that area, probably
isn't doing a very good job...

Also...Joe noted that your coax feedline may well be a length that
accounts for the SWR peaks and valleys. (I think it may be about twice
as long as Joe wrote...but same idea.) Do you see the peaks and
valleys when you terminate the line in the precision 2:1 load? If you
do NOT, then it's a further indication that the feedline has antenna
currents on it, because the flat 2:1 is an indication that your
transmission line is matched to the calibration impedance of the SWR
bridge, and if that's the case, the SWR bridge should be giving at
reasonably accurate estimate of the actual line SWR. If you DO see the
SWR ripples vs frequency with just the precision load, either the load
isn't "flat" or the line is not the same impedance as the SWR bridge is
calibrated to, and the differing impedances is by far the most probable
explanation if the line length is right.

Cheers,
Tom

I would expect the outer is a Teflon or Teflon-like material, with
probably low loss and fairly low relative dielectric constant, but it's
still going to require that you shorten that lower section quite a bit
to get the quarter-wave stub to reflect a high impedance. In addition,
the dielectric between the outer of the coax and the inner of the
sleeve is pretty thin, and the impedance of the resulting coaxial
arrangement is pretty low. That means that it won't ever reflect a
very high impedance. You can use a tapered sleeve that bells out at
the bottom, to good effect. That will also lower the feedpoint
impedance and match better to 50 ohms. Think: ground plane with
drooping radials. But you can also wind the coax just below the
antenna into a small coil (I'd make the axis of the coil coincide with
the axis of the antenna) that's self-resonant near your operating
frequency, and it will very effectively choke off antenna current
(current on the outside of the coax) at that point. Put one such coil
an inch or so below the bottom of the sleeve, and another about a
quarter wave further down the line.

SWR meter: Get RFSim99--do a Google search for it. Build a coupler,
per toolsdesigncoupler. I'd suggest a microstrip version, if you
can make a little PC board reasonably accurately. Design it for 50
ohms. Terminate each coupled port in 50 ohms (e.g. 49.9 ohm 0805 SMT
part). Using vanishingly short leads, connect a simple diode detector
to each of those two loads. Use a calibrated attenuator to calibrate
at least the relative response of those detectors. Use those two
outputs to calculate SWR. You can read the diode detector outputs with
a DVM that has good resolution (10uV sensitivity preferred; 1uV is even
better). I'd recommend about a 20dB coupler for the power level you're
using, though even a 30dB coupler would work. Try terminating the
through line in a 49.9 ohm (or a parallel pair of 100 ohm) 0805 parts
to check that you see essentially no reflected, and try a 100 ohm load
and a 25 ohm load to check that you get the expected reflected. -- To
have the coupling right according to the RFSim99 directions, it needs
to be 1/4 wave long, but it's a pretty broad peak. Coupling drops to
zero at 1/2 wave, and at DC. So you could make one for 900MHz, and it
would work OK at 450MHz, you'd just get lower coupling. The ratio for
SWR would still be OK. The 450MHz version on FR4 board
(fiberglass-epoxy) would be roughly four inches long, if my mental
arithmetic is right, and half that for 900MHz.

Cheers,
Tom


David wrote:
Tom,

Thanks for the information. The inner coax is the smaller RG174 coax.
The sleeve is made of earth braid pulled from RG58 cable. The dielectric
between the sleeve and inner cable is therefore the outer sheath of the
RG174 cable (Not sure what this is, the RG174 I have is Teflon inner
dielectric and stranded conductor. The utter sheath is a very strong
heat resistant material - I therefore have no ideal of the dielectric
constant to calculate Vp for correct electrical length). If I use a
copper tube and strip off the sheath from the inner coax, I can
calculate correct length as it will have an air dielectric.

Do you know where everyone is getting the dielectric constants for
various materials ? I noted people using small metal tubes as sleeves
and quoting these magic numbers even for copper tubes of certain diameters.

From my discussion with Telonic, they say the Rho_Tector was designed
as an in-house tool for measuring inputs of amps and filters. They
suggest SWR meter would probably be best for antenna adjustments.
Do you happen to know where I might find details for a low power SWR
meter for 915 MHz ? I need one that will operate with only 20mW applied
power. The only SWR meter I have has min. FSD of 3W

Thanks

Regards

David

K7ITM wrote:
Hi David,

You wrote, "The decoupling is via a 1/4 wave sleeve that provides high
impedance for
RF returning along outer coax and also as the second 1/2 of the
dipole. "

EXACTLY how is this built? The details of construction make a BIG
difference in performance! (There's a lot of BAD info about it out
there...)

It's not a bad idea to ALSO put some additional decoupling further down
the feedline.

If your spectrum analyzer/field strength meter is far enough away from
the antenna you are testing, then it should provide a reasonable
indication of relative antenna radiation performance. The SWR
indication, if properly calibrated and given that you are apparently
exciting the antenna with a source whose output impedance matches your
feedline, should also be a good indication of power actually absorbed
by the antenna. That is, lowest SWR represents maximum power absorbed
by the antenna. Presumably that power is being radiated as RF, mostly,
and not dissipated as heat. But where the RF radiation goes depends on
the pattern of currents excited on the conductors that compose the
antenna, and nearby conductors as well (such as the feedline). What
you probably want is standard resonant half-wave dipole currents on
your vertical dipole, and no (very little) antenna current on the
feedline and on support structures. By the way, whether the antenna is
resonant or not is of little real importance, so long as you can
efficiently feed power to it and the antenna currents are in the right
places and not the wrong places. But it happens that with your
antenna, if things are working properly (properly decoupled feedline,
etc), you probably will see lowest SWR at half-wave resonance. If you
have no other matching going on, the lowest SWR will probably be about
1.5:1 with 50 ohm feedline. You could add parts to get a better match
if you wished.

And as you can probably tell from all that, I'm suspecting that your
decoupling sleeve, with associated dielectrics in that area, probably
isn't doing a very good job...

Also...Joe noted that your coax feedline may well be a length that
accounts for the SWR peaks and valleys. (I think it may be about twice
as long as Joe wrote...but same idea.) Do you see the peaks and
valleys when you terminate the line in the precision 2:1 load? If you
do NOT, then it's a further indication that the feedline has antenna
currents on it, because the flat 2:1 is an indication that your
transmission line is matched to the calibration impedance of the SWR
bridge, and if that's the case, the SWR bridge should be giving at
reasonably accurate estimate of the actual line SWR. If you DO see the
SWR ripples vs frequency with just the precision load, either the load
isn't "flat" or the line is not the same impedance as the SWR bridge is
calibrated to, and the differing impedances is by far the most probable
explanation if the line length is right.

Cheers,
Tom