SWR Analyzer VS. SWR Meter
Robert Lay
Does anyone have an explanation for the apparent disagreement between
SWR Analyzer and SWR Meter in the following situation:
- Antenna is G5RV in inverted vee config.
- Coax feedline has outer braid earthed by coax switch at entry point to
shack.
- Antenna coax connected to MFJ-989C Tuner at Antenna port.
- Configuration "A": SWR Analyzer (by MFJ) connected at transmitter side
of tuner to determine when tuner is providing proper match.
- Configuration "B": SWR Meter in Tuner used to determine proper match
with transmitter connected to transmitter side of tuner.
If the tuner is tuned to give an apparent match in Configuration "A",
and then the transmitter is hooked up in Configuration "B" and the tuner
is re-tuned to give proper match according to internal SWR meter, the
tuner settings are noticeably different.
Additional observations: This situation obtains with some antennas and
not with others, but when this situation does occur, it is always
accompanied by hiher degree of interaction between the tuner controls.
I am suspicious that it is due to RF flowing on outer braid of coax.
How might one determine whether or not that is the case?
Has anyone else seen this problem?
73
Bob
N8BLK
In article <33D0E1...@crosslink.net>, Robert Lay <w9...@crosslink.net>
writes:
>If the tuner is tuned to give an apparent match in Configuration "A",
>and then the transmitter is hooked up in Configuration "B" and the tuner
>is re-tuned to give proper match according to internal SWR meter, the
>tuner settings are noticeably different.
>
>Additional observations: This situation obtains with some antennas and
>not with others, but when this situation does occur, it is always
>accompanied by hiher degree of interaction between the tuner controls.
>I am suspicious that it is due to RF flowing on outer braid of coax.
>
>How might one determine whether or not that is the case?
>Has anyone else seen this proble
Easiest way to test which is giving accurate information on SWR is to
either compare with a known accurate power meter, which gives forward
and reflected readings, and compute the swr. You can also of course
measure the swr readings of both at the frequency in question using a
dummy load. One other possiblity is that something in the tuner
breaks down under an RF load, but the MFJ doesn't put out enough rf to
produce the same problem..
Another is that you're trying to use the MFJ on 160 meters, and are close
to a medium wave station transmitter, the rf from the radio station can't
be
distinguished from the normal rf coming from the MFJ so you get
inaccurate readings.
-art clemons-
Dale Martin
Bob,
This is something I have wondered about, too. Just today my MFJ-259 was
showing
about a 2:1 SWR on 14.0/21.0/28.0 MHz on our triband beam at our club
station.
When we put a radio on the antenna, we were seeing better than 3.5:1 SWR on
the
three bands. Obviously, there's a problem with the feedline or the
antenna.
When I got home, I checked the analyzer on my own antenna (same as our
club's)
and it's readings across the three bands closely matched those on the
SWR/Watt
meter with the transmitter.
But, what's to say that there is not a variation in each piece of equipment
you are
working with such that the impedance of each is actually different than
what it is
supposed to be (50 ohms, nominal)? Maybe your analyzer does not have or
exhibit the same impedance as your transmitter.
One way to check for consistency of the analyzer might be to set up
configuration
B, then take and record readings with configuration A. Change the tuner
settings
and see if you can duplicate the configuration B settings by adjusting the
tuner
with the 'new' or 'compensated' settings given by the analyzer.
Good luck,
--
Dale Martin
kg...@hal-pc.org
http://www.hal-pc.org/~kg5u
Robert Lay <w9...@crosslink.net> wrote in article
<33D0E1...@crosslink.net>...
> Does anyone have an explanation for the apparent disagreement between
> SWR Analyzer and SWR Meter in the following situation:
> - Antenna is G5RV in inverted vee config.
> - Coax feedline has outer braid earthed by coax switch at entry point to
> shack.
> - Antenna coax connected to MFJ-989C Tuner at Antenna port.
> - Configuration "A": SWR Analyzer (by MFJ) connected at transmitter side
> of tuner to determine when tuner is providing proper match.
> - Configuration "B": SWR Meter in Tuner used to determine proper match
> with transmitter connected to transmitter side of tuner.
>
> If the tuner is tuned to give an apparent match in Configuration "A",
> and then the transmitter is hooked up in Configuration "B" and the tuner
> is re-tuned to give proper match according to internal SWR meter, the
> tuner settings are noticeably different.
>
> Additional observations: This situation obtains with some antennas and
> not with others, but when this situation does occur, it is always
> accompanied by hiher degree of interaction between the tuner controls.
> I am suspicious that it is due to RF flowing on outer braid of coax.
>
> How might one determine whether or not that is the case?
> Has anyone else seen this problem?
> 73
> Bob
>
Joe FITR BV/N0IAT
Be careful if you are using the MFJ analyzers. I have noticed
the analyzers become INACCURATE when the antenna is in the
presence of strong RF (from a nearby BC station, other ham
station, etc). Guess it's RF from the near-field that is
getting into the analyzer and causing false readings.
Joe
Robert Lay
phenomenon is repeatable any time of day or night, and my QTH is 30
miles from the nearest broadcast station, which is on the FM band. My
measurements that manifest this problem are typically on 80 through 20
meters - certainly never on any frequency lower than 3.5MHz.
I'm still listening - and working the problem.
Bob
Robert Lay
Dear Art,
I decided to conduct some tests designed specifically to evaluate the
effects of having my coax lines go through the grounding switchblock as
they come through the shack wall from the outside. Each coax shield
becomes earthed at this point by virtue of a bank of s.p.d.t coax
switches which all connect to an aluminum plate having a short pigtail
to a ground rod just outside the shack.
I brought the coax from the G5RV through the wall with no grounding and
straight to the tuner and MFJ SWR Analyzer for this test. Under those
conditions the SWR Analyzer and the tuner's SWR meter give much better
agreement and the tuning controls of the tuner are not so interactive.
As soon as I go back to my original configuration of having an earth
ground on the coax shield at the entry point to the shack the tuning
controls are more interactive, the tuner settings for a match shift to
different point to get a match and the MFJ SWR analyzer no longer agrees
with the tuner's SWR meter. (This was all done in the middle of the 75 M
phone band.)
This doesn't satisfy me completely as to what's going on, but it does
make me believe that 1) there is considerable RF on the outside of the
shield and 2) the G5RV is probably more prone to this kind of problem
than is an antenna with a low (1.3:1 or lower) SWR on its coax feed.
It also re-enforces my belief that the antenna tuners should be at the
other end of the coax feed - not in the shack - but what the hell are
you gonna do?
73 and thanks again,
Bob
James C. Marciniak
Robert Lay wrote:
>
> Does anyone have an explanation for the apparent disagreement between
> SWR Analyzer and SWR Meter in the following situation:
> - Antenna is G5RV in inverted vee config.
> - Coax feedline has outer braid earthed by coax switch at entry point to
> shack.
> - Antenna coax connected to MFJ-989C Tuner at Antenna port.
> - Configuration "A": SWR Analyzer (by MFJ) connected at transmitter side
> of tuner to determine when tuner is providing proper match.
> - Configuration "B": SWR Meter in Tuner used to determine proper match
> with transmitter connected to transmitter side of tuner.
>
> If the tuner is tuned to give an apparent match in Configuration "A",
> and then the transmitter is hooked up in Configuration "B" and the tuner
> is re-tuned to give proper match according to internal SWR meter, the
> tuner settings are noticeably different.
>
> Additional observations: This situation obtains with some antennas and
> not with others, but when this situation does occur, it is always
> accompanied by hiher degree of interaction between the tuner controls.
> I am suspicious that it is due to RF flowing on outer braid of coax.
>
> How might one determine whether or not that is the case?
> Has anyone else seen this problem?
> 73
I had some supposedly non-inductive 25-watt 25 ohm resistors, and some 50
watt 100 ohm resistors, and I wanted to see just how "non-inductive" they
really were. I used an MFJ analyzer, and bolted the resistors together
such that the analyzer would see a 50 ohm load.
I connected the resistors to the MFJ using a 12 inch lencth of RG-8
cable, with a PL-259 at one end, and "flea clip" terminations on the
other (flea clips to the resistors).
What I saw was pretty wild, as I ran the frequency from 5 - 30
MHz...Impedance was certainly not flat! the Z varied from 50 Ohms at 5
mHz to 99 at 30 MHz!
A friend tried the same thing using the Autek analyzer, and got a
significantly flatter result (although it was obvious these were not NI
resistors at shortwave frequencies!) His curve was 50 ohms at 5 MHz, 74
at 30 MHz, and much more linear - my curve looked like the matterhorn! I
was sure the MFJ was at fault...
Well, it turns out that 12 inch piece of coax was creating the problem!
using a banana jack-flea clip connection to the center of the MFJ's
SO-238 connector, with a back-to-back flea clip on the ground portion of
the connector resulted in repeatable readings between the Autek and the
MFJ analyzers. Autek recommends this type of test setup for the
application I was using, while MFJ had nothing in its manual - other than
a vague warning that loads that become reactive will not read properly on
the meter.
I also think their math is wrong concerning SWR measurement - at least it
conflicts with Autek's way of doing things...
--
"Some scientists claim that hydrogen, because it is so plentiful,is the
basic building block of the universe. I dispute that. I say there is more
stupidity than hydrogen,and that is the basic building block of the
universe."
- Frank Zappa
Tom Rauch
>Be careful if you are using the MFJ analyzers. I have noticed
>the analyzers become INACCURATE when the antenna is in the
>presence of strong RF (from a nearby BC station, other ham
>station, etc). Guess it's RF from the near-field that is
>getting into the analyzer and causing false readings.
>
>Joe
Joe is correct, ALL analyzers of this or similar types (using a
UNtuned detector in the bridge) are susceptable to externally induced
power on the feedline, whether the interference is CLOSE to the
transmitted frequency or not. As little as a 1/4 volt of external
signal will skew the SWR readings, always making them HIGHER than they
really are on the Analyzer.
Outside interference from close-by sources will NEVER make the 259 or
other analyzers (the Autek) read LOWER than the SWR really is.
But in checking the tuner, the fellow noticed a change in SWR from the
inside bridge to the MFJ. In that case, I'd trust the MFJ.
The MFJ uses fixed resistors in a convenmtional bridge. While there
are SWR level adjustments, there are NO 50 ohm null adjustments. A
conventional directional coupler, like used in the antenna tuner,
usually has a trimmer capacitor for a null adjustment. If someone
adjusted the trimmer incorrectly, the bridge could null at some
impedance other than 50 ohms. Defects in design can also cause the
bridge to null with reactive loads, making you think you have a 1:1
SWR when it is high.
Many bridges are mounted in tuners where the RF field or chassis
common mode currents affect the bridge null. So a built in tuner SWR
bridge that nulls great on some loads may null poorly on other
loads!!!
It sure sounds like bridge trouble in the tuner to me.
73 Tom
Tom Rauch
>I also think their math is wrong concerning SWR measurement - at least it
>conflicts with Autek's way of doing things...
What math did you find wrong or question?
>"Some scientists claim that hydrogen, because it is so plentiful,is the
>basic building block of the universe. I dispute that. I say there is more
>stupidity than hydrogen,and that is the basic building block of the
>universe."
> - Frank Zappa
That statement is supported by the fact Zappa wanted to spend his life
on a farm trying to grow dental floss. ;-)
73 Tom
Tom Rauch
On Mon, 21 Jul 1997 17:47:15 +0100, Robert Lay <w9...@crosslink.net>
wrote:
>I brought the coax from the G5RV through the wall with no grounding and
>straight to the tuner and MFJ SWR Analyzer for this test. Under those
>conditions the SWR Analyzer and the tuner's SWR meter give much better
>agreement and the tuning controls of the tuner are not so interactive.
>As soon as I go back to my original configuration of having an earth
>ground on the coax shield at the entry point to the shack the tuning
>controls are more interactive, the tuner settings for a match shift to
>different point to get a match and the MFJ SWR analyzer no longer agrees
>with the tuner's SWR meter. (This was all done in the middle of the 75 M
>phone band.)
Hi Bob,
Common mode current problems can cause the effect you describe, but
it is not SWR related.
It is because the G5RV is a balanced antenna fed by an unbalanced
transmission line.
I'd use a choke type balun near the antenna, you can make one by
winding ten or fifteen turns of coax in a nice single layer on a four
inch PVC drain pipe or something similar.
I posted a reply about the meter in the tuner possibly being off, but
after reading you latest post it sounds more like common mode
problems. It also sounds like you do NOT have the case of the tuner
grounded to a good ground. That is not a good idea, I'd ground the
tuner to the station ground.
>It also re-enforces my belief that the antenna tuners should be at the
>other end of the coax feed - not in the shack - but what the hell are
>you gonna do?
Naw, you just need to clean up the common mode current unless the
feedline is very long or very lossy. Moving the tuner is nice, but
tuff to do unless you never change bands. Besides, moving the tuner
won't fix the common mode current problem.
73 Tom
Robert Lay
Glad you got in on the dialogue, Tom. The tuner that I am using is the
MFJ-989C, which I am very happy with. Especially since its twin needle
Power and SWR meter always seems to agree with other Power/SWR meters
that I put in cascade with it (i.e., between the transmitter and the
MFJ-989C).
I am also grateful for the additional insights into the construction of
my MFJ SWR Analyzer.
Do you have any additional comments on the situations where the tuner
controls become so painfully interactive? I've felt for some time that
it must be due to RF flowing on the outer braid, but haven't seen
anything but conjecture on that point.
Grateful,
Bob
Tom Rauch
On Tue, 22 Jul 1997 22:07:46 +0100, Robert Lay <w9...@crosslink.net>
wrote:
>
>Dear Tom,
>Glad you got in on the dialogue, Tom. The tuner that I am using is the
>MFJ-989C, which I am very happy with. Especially since its twin needle
>Power and SWR meter always seems to agree with other Power/SWR meters
>that I put in cascade with it (i.e., between the transmitter and the
>MFJ-989C).
Hi Bob,
Actually the answers everyone posted were correct, the problem is
almost always one of picking through all the various causes and
finding the most likely cause.
>Do you have any additional comments on the situations where the tuner
>controls become so painfully interactive? I've felt for some time that
>it must be due to RF flowing on the outer braid, but haven't seen
>anything but conjecture on that point.
You could build a clamp on meter, and measure the common mode. I think
both K1BQT and W6RCA, who post on this newsgroup, have made practical
common mode measurement gizmos. I'm sure others will have suggestions.
But just the fact your second description indicated a change in SWR
measurements when a grounding block was added is a very strong
indicator common mode currents are a large part or all of the problem.
73 Tom