Should feedline lenght change the VSWR?
Cecil_...@ccm.ch.intel.com
>So I called Company "G" and spoke to
>their tech support person and he told me that one should not
>measure the VSWR at the antenna.
He is probably right. Your analyzer is probably affected by the near
field of the antenna and will give bad readings especially if it has a
microcomputer in it. (I'm a microcontroller jock).
>and that the VSWR would change depending on the length of the feed line.
He's right again. No matter what the SWR is at the antenna, if you
make your coax long enough, the SWR at the generator will be close
to 1/1.
>This isn't how I understood things worked from reading Walter Maxwell's
>book "Reflections" but maybe I missed something.
Never, never, never assume that coax is lossless no matter what anybody
says. Learn when transmission line losses matter and when they don't.
Every transmission line, no matter how long or short and no matter
how high the quality, has losses!!! Even 100 ft of 450 ohm ladder-line
has measurable losses on 80m (but much, much less than coax).
On 440 MHz with 200 ft of RG-58, your SWR will be less than 1.1/1
with no antenna attached or with the coax shorted at the load end.
I don't know a thing about that particular antenna but sounds like
they are depending on the losses in the coax and the transformer
characteristics of the transmission line to give a reasonable match
at the generator.
73 and good luck, Cecil, KG7BK, OOTC (Not speaking for Intel)
--
Intel, Corp.
5000 W. Chandler Blvd.
Chandler, AZ 85226
Cecil_...@ccm.ch.intel.com
>Thanks for taking the time to read this and I will appreciate
>any and all feed back. >Terry KJ7F
Hi Terry, I'm back... didn't do a good job of explaining concepts in my
last posting. Here's a true/false test for everybody interested in
antennas and transmission lines.
1. T/F There is such a thing in reality as an SWR of exactly 1/1.
2. T/F An open-circuited or short-circuited transmission line has an
infinite SWR at the load end.
3. T/F The SWR doesn't change from the load end to the generator end
of a transmission line.
4. T/F An SWR meter does not change the SWR.
All are false. There is no such thing as an antenna that perfectly matches
the characteristic impedence of a transmission line, i.e. 50.00---00 +
j0.00---00 does not exist. _For practical purposes_ there is such
a thing as an SWR of 1/1. In reality, there is also no such thing as
a precision resistor of exactly 50 ohms, just one as close as you want.
An open-circuited transmission line does not have an infinite impedence.
A short-circuited transmission line does not have a zero impedence. _For
practical purposes_ it may be assumed to be true under certain conditions.
Everyone should learn those conditions.
The SWR changes every inch up and down the transmission line because every
inch has a finite loss and because the SWR is never exactly 1/1 at the
load. _For practical purposes_ if the transmission line is matched
to the load the SWR is close enough to 1/1. If the transmission line is
not matched to the load, the SWR decreases in the direction toward the
generator because of losses. Removing an inch of transmission line
changes the SWR... inserting an SWR meter changes the SWR... hopefully
by a neglible amount.
SWR's of 1/1, infinite and zero impedences, exactly 50 ohm antennas and
transmission lines are human concepts that do not exist in reality.
Sometimes they are good approximations to reality and sometimes they are
not.
Everyone who does not understand when these items are relevant and when
they are not should study and understand the graphs at the end of the
Transmission Lines chapter in the ARRL Handbook. When Maxwell says
"assuming no losses in the transmission line" he really means that the
losses in the transmission line are neglible for what he is trying to
explain. In reality, there is no such thing as "no losses". The load
never receives all of the generator's power.
Assume a 50 ohm antenna and one electrical wavelength of 450 ohm ladder-
line. What is the SWR? ... close to 9/1. What impedence does the
generator (transmitter) see? ... close to 50 ohms. You will deliver more
power to the antenna with this configuration than you will with RG-58
coax and an SWR of 1/1... believe it or not.
73, Cecil, KG7BK, OOTC (Not speaking for Intel)
Micron3
question about SWR that
I have. I recently bought a new vertical antenna which I installed on
the roof of my house.
I is one of the center fed type of verticals that isn't supposed to need a
ground plane that
Company "G" advertizes quite widely. After I installed it I check the
VSWR in the shack with
my VSWR bridge and found that the VSWR on most of the bands was rather
high. In most cases
between 2 to 1 and 4 to 1. I then went up to the base of the antenna
with my Autek RF Analyst
and measure the VSWR and found that all of the bands had a very high SWR.
For example,
20 Meters was 4.1 at the bottom of the band and 5.0 at the top end. 75
meters was the best
with a reading of 1.6 to 1 near the top of the band. So I called Company
"G" and spoke to
their tech support person and he told me that one should not measure the
VSWR at the antenna.
65 feet long and that
the VSWR would change depending on the lenght of the feed line. I told
him that I was under
the impression that the VSWR on the feed line should not change with
lenght unless the load,
in this case the antenna were unbalanced. He launched into a long
lecture about how this was
only true if the load presented a perfect 50 Ohm non reactive resistance
and that in the real
world of an antenna this condition could not be met and therefor the VSWR
would change as you
change the length of the feed line. This isn't how I understood things
worked from reading
Walter Maxwell's book "Reflections" but maybe I missed something. It
wouldn't be the first
time! Then he went on to tell me that the VSWR on 75 meters should be
lower than 1.6 and that
the rest of the antenna wouldn't work unless the VSWR on 75 was lower.
He asked if I had
any other antennas near the vertical and I said that I had a 40 foot
frestanding tower about 70
feet away and he told me that the tower was messing up the vertical
because it was less than
one wavelenth away! Any one have any ideas about that one?? I want to
call him back and
further our discussion but I want to make sure I have my facts straight
first. Is what this
guy told me correct or is he just putting out a smoke screen? Thanks for
Micron3
writes:
>Your analyzer is probably affected by the near
>field of the antenna and will give bad readings especially if it has a
>microcomputer in it. (I'm a microcontroller jock).
I am curious why the fact that the RF Analyst has a
microprocessor in it would affect the accuracy?
>He's right again. No matter what the SWR is at the antenna, if you
>make your coax long enough, the SWR at the generator will be close
>to 1/1.
>The SWR changes every inch up and down the transmission line because
every
>inch has a finite loss and because the SWR is never exactly 1/1 at the
>load. _For practical purposes_ if the transmission line is matched
>to the load the SWR is close enough to 1/1. If the transmission line is
>not matched to the load, the SWR decreases in the direction toward the
>generator because of losses. Removing an inch of transmission line
>changes the SWR... inserting an SWR meter changes the SWR... hopefully
>by a neglible amount.
I agree with you that there will be some loss in the coax and that
it will cause the VSWR reading in the shack to be lower than at the
antenna. I guess I didn't word that part very well. What I was
trying to understand is the fact that I read a VSWR at the antenna
on 14.200 MHz of 4.3 - 1 and in the shack it reads 1.5 while on
21.200 I read 2,1 - 1 at the antenna and 2.4 - 1 in the shack?
The gentleman at the store told me this is normal and I need to cut
the coax feed to 65 feet and then the VSWR curves would fall in
place as the VSWR varies up and down with length. At least that
is what I thought he said.
Thanks for taking the time to reply, it is appreciated!
Cheers, Terry KJ7F
These are not the opinions of my employeer, I have enough trouble having
my own!
Cecil_...@ccm.ch.intel.com
>
> I am curious why the fact that the RF Analyst has a
>microprocessor in it would affect the accuracy?
Hi again, Terry. There are millions of digital bits roaming around
inside a microcomputer. RF fields can change those bits. Somewhere
inside the device they probably do an analog to digital conversion
and a one bit error can be a 50 percent error. I would not trust any
readings done with ordinary ham-grade test equipment in the presence
of the antenna near field and especially not anything containing a
microcomputer. Away from EM fields, it should operate just fine.
>What I was
>trying to understand is the fact that I read a VSWR at the antenna
>on 14.200 MHz of 4.3 - 1 and in the shack it reads 1.5 while on
>21.200 I read 2,1 - 1 at the antenna and 2.4 - 1 in the shack?
It's likely that the characteristic impedence at the antenna is not 50
ohms and remember your SWR meter is standardized for 50 ohms. If
the characteristic impedence at the point of measurement is not the
same as your SWR meter, results may look wierd. For instance, in my
earlier example of a 50 ohm antenna fed with one electrical wavelength
of 450 ohm transmission line and an SWR of 9/1, a 50 ohm SWR meter at
the generator will not read 9/1. You would need a 450 ohm SWR meter.
>The gentleman at the store told me this is normal and I need to cut
>the coax feed to 65 feet and then the VSWR curves would fall in
>place as the VSWR varies up and down with length. Cheers, Terry KJ7F
Maybe you could work this thing backwards. By measuring the SWR and
impdeance at the generator end and knowing the length and matched-line
loss of the transmission line, you can calculate the SWR and the
impedance at the antenna. The adjustments on these multi-band verticals
interact so that changing the SWR on one band will also change the SWR
on the other bands. I would be tempted to feed it with ladder-line and
an antenna tuner and forget the SWR.
73 and good luck, Cecil, KG7BK, OOTC (Not speaking for Intel)
Stephen Modena
the text wrap-around value....
I've seen several of Cecil's comments...he is entitled to his opinion,
but I don't agree with his approach...
I'm surprised that you're afraid to mention the make of the antenna,
but didn't hesitate to mention Autek...I can't even look in a catalog
for some guidance...that's a shame.
Now, let's try to sort it out:
By *definition* the antenna you have is *unbalanced*....it
may be center-fed and not need "radials", but it is mounted vertically
and it is definitely unbalanced with respect to ground.
How unbalanced it is depends on how close it is to true earth
ground and to anything else that might interact with it at RF freqs
via capacitive coupling...and if it is mounted on your house, there is
bound to be something...so the question is the *degree* of unbalance.
Let's recall another source of antenna "unbalancing"...the use of
coax. Unless there is sufficient choking on the outer braid surface
at the antenna feed point...and at other points leading away, then
your antenna is a TRIpole, not a dipole...and by definition, unbalanced.
Why did I say, farther away along the coax? Because if you do not
lead the coax away from the antenna so that the outer braid surface
is *symmetrically* located with respect to *both* arms of the antenna,
the antenna is by definition unbalance by virtual of coupling.
Let's look at another potential source of unbalancing: MFJ's and
others appear to use a fairly long "upper" element, but the "lower"
half of the vertical dipole is quite short via inductive loading...
That antenna may be very, very difficult to keep in "balanced mode" at
all but one frequency on each band....because the inductively loaded
half does not have the same reactance-vs-freq sweep curve. In other words,
you should take it that the antenna is *always* unbalanced, and by
quite a bit.
But frankly, I do not believe that this the central problem.
I will note that you do not seem to have told us anything about how
you tuned up this antenna on each band after you assembled it. That
could be very important.
Now...let's turn to the idea that changing the length of the transmission
line would change the SWR...well yes of course, in your case...it is
almost unimaginable that the feed point of your antenna is 50 ohms
resistive on any but a single frequency...but in fact, is probably
never anywhere near 50 ohms over most of its operation range...and
who knows what the reactive component is....
It follows that a mismatch between the antenna
feed point impedance and the surge (characteristic) impedance of the
transmission line will result in the *transmission line* acting
as an impedance *transformer*...and by definition, the nature of the
impedance transformation is *dependant* the *length* of the coax.
This stuff is in the ARRL Handbook briefly and in more depth in the
Antenna Mannual.
Depending on the actual degree of mismatch *and* the absolute values
involved, it is *not* unusual for SWR values to *soar* when the
transmission line length is at the *sharply* climbing regions of the
Tan(h) function of impedance transformation. Solution? Lengthen or
shorten the transmission line...or put an appropriate open-or-shorted
shut about 0.1 wavelength from the antenna feed point to make the
transmission line go flat...solutions that are easy on a single band
but may require considerable experimentation when frequency hopping
multiple bands is involved (as is the case of your multi-band antenna).
Now let's visit what the factory rep was trying to get across to you.
Why not measure at the antenna (feedpoint)?
Because you are interacting with the antenna by being there...because
if you are interested in taming SWR with a noise bridge, reading
at the antenna will not tell (inexperienced) you very much that *you*
know how to *use*....this is not an insult: your post makes it clear
that you do not have enough understanding to use at-antenna readings
to advantage....that might not be the case for Roy Lewellen or me,
but you are the one struggling along...hence it is better to watch
the SWR at the point where it plugs into your transmitter when
adjusting the antenna.
The factory rep says that you should be using 67 feet of coax to make
measurements...I'd strongly recommend that you do as you are instructed
so that the *rep* will have a set of readings in a frame of reference
from which he can help you. Be sure to use the *particular* type of coax
he recommends...especially since there is a big difference between
RG-8 and RG-58 in the resulting readings.
Failure to follow factory recommendations in trouble shooting means that
your conversations with him are *probably* a waste of *his* time.
The factory rep is 1,000 miles from your antenna...he may have a
standardized procedure for analyzing problems *by telephone*...and
I'd advise against rejecting his *requirements* in order to help you.
Let me illustrate: we used to run a foreign car repair shop...and
as a condition of diagnosing engine or running problems...we
required the customer agree to *us* putting is a new set of spark
plugs...that component is so critical and so *easily* fouled in
certain cases, that it was a sine-qua-non. Sometimes the customer
would say: I don't think it's the plugs...and we would say, if you
already know what the answer is or isn't, why do you need us? The
point is: we have never seen the car before, don't know what has
been done or happened to it...and we can not mind read the car or the
customer...so you develop certain solid preconditions to prevent
sabotage of the diagnosis *process*.
Your factory rep is trying to help you...let him: do what he wants
and be prepared to put in a lot of effort to meet his diagnostic
requirements...he is, after all, 1,000 miles away and doing this
by remote control. If you can fix this by yourself or via the net,
then you don't need to listen to the factory rep, right? :^)
Remember: neither the factory rep nor us on the net, know what you've
done to and with that antenna, whether you know how to solder a
coax connector correctly...or whether you've got rain water in
your coax! ;^) Good luck and keep us posted.
73/Steve/AB4EL
--
Steve Modena AB4EL ** National Radio Examiners--Test Center Manager
ab...@Cybernetics.NET ** We offer periodic Commercial Radio Operator
** license examinations in the RTP, NC area.
John F. Woods
>>not matched to the load, the SWR decreases in the direction toward the
>>generator because of losses. Removing an inch of transmission line
>>changes the SWR... inserting an SWR meter changes the SWR... hopefully
>>by a neglible amount.
> I agree with you that there will be some loss in the coax and that
>it will cause the VSWR reading in the shack to be lower than at the
>antenna. I guess I didn't word that part very well. What I was
>trying to understand is the fact that I read a VSWR at the antenna
>on 14.200 MHz of 4.3 - 1 and in the shack it reads 1.5 while on
>21.200 I read 2,1 - 1 at the antenna and 2.4 - 1 in the shack?
>The gentleman at the store told me this is normal and I need to cut
>the coax feed to 65 feet and then the VSWR curves would fall in
>place as the VSWR varies up and down with length. At least that
>is what I thought he said.
Never, ever believe what "the gentleman at the store" says. Regardless
of the store.
If your antenna is being fed with 50-ohm coax, under no circumstance does
the SWR change (noticably) with the length of coax. If the coax between
the meter and the antenna is *not* 50 ohms, then it's acting as an impedance
transformer; but this is almost certainly not the case, especially since
the "helpful" salesman is telling you to use a quarter wavelength.
If you have actually changed the length of the coax and seen the SWR change
dramatically, you have RF current flowing on the outside of the coax (and
thus have a markedly varying amount of return power being seen by the bridge).
Use some kind of choke or balun at the antenna to decouple the coax and you
will get more repeatable readings. If the SWR meter accurately measures
an SWR of 4.3:1 at the antenna, then that is absolutely what the SWR is,
and you want to trim the antenna to resonance. (Or get a decent wide-range
tuner, feed the antenna with openwire, and LAUGH at high SWR readings.)
John F. Woods
>It follows that a mismatch between the antenna
>feed point impedance and the surge (characteristic) impedance of the
>transmission line will result in the *transmission line* acting
>as an impedance *transformer*...and by definition, the nature of the
>impedance transformation is *dependant* the *length* of the coax.
>This stuff is in the ARRL Handbook briefly and in more depth in the
>Antenna Mannual.
Correct.
>Depending on the actual degree of mismatch *and* the absolute values
>involved, it is *not* unusual for SWR values to *soar*
False. Given a lossless line, the SWR is constant along the entire line.
The impedance transformation performed by a length of cable always results
in the same SWR (if you graph the results on a Smith chart, you end up rotating
the load impedance along a circle of constant radius centered on the
characteristic impedance of the line).
Let us work out some examples, using the formulae from the Handbook. Let
the coax be 50+0j ohms and the load be 100+0j ohms. For line lengths of
33 degrees, 45 degrees, 80 degrees, 90 degrees, 150 degrees, and 180 degrees,
the impedances (and SWR values) are:
33 degrees Ra 52.912992 Xa -36.253817 SWR 2.000000
45 degrees Ra 40.000000 Xa -30.000000 SWR 2.000000
80 degrees Ra 25.578464 Xa -6.561262 SWR 2.000000
90 degrees Ra 25.000000 Xa -0.000000 SWR 2.000000
150 degrees Ra 57.142857 Xa 37.115374 SWR 2.000000
180 degrees Ra 100.000000 Xa 0.000000 SWR 2.000000
Note the extremely boring nature of the last column. If you examine the
cases for 90 degrees and 180 degress, you find the expected cases (half
wavelength repeats the input impedance, quarter wavelength follows
Rin = (Zo^2 / Ra)) doing the expected things, and the intermediate cases
should argue for continuity.
The program used to calculate the above numbers follows; compare it against
the formulas in the Handbook.
#include <math.h>
/* notes on equations: ra is resistive component of load impedance, normalized
* to characteristic impedance of line (taken as 1.0). xa is normalized
* reactive component of load impedance. Bl is "beta script-l" from the
* equations as printed, and represents the electrical length of the line
* (here in radians, rather than degrees). "tan Bl" in the comments represents
* tan(bl) in the code.
* The calculations are broken up into pieces since I don't trust the
* C compiler to do common subexpression elimination well enough; on the
* other hand, why did I waste my time if speed isn't critical? Geez.
*/
/* 2
* ra (1 + tan Bl)
* r = ----------------------------------
* in 2 2
* (1 - xa tan Bl) + ( ra tan Bl )
*/
double rin(bl, ra, xa) /* bl in radians, ra/xa in normalized ohms */
double bl, ra, xa;
{
double tanbl = tan(bl);
double num = ra * (1 + tanbl * tanbl);
double denom1 = (1 - xa * tanbl);
double denom2 = (ra * tanbl);
return num / (denom1 * denom1 + denom2 * denom2);
}
/*
* 2 2 2
* xa (1 - tan Bl) + (1 - ra - xa ) tan Bl
* x = ------------------------------------------
* in 2 2
* (1 - xa tan Bl) + (ra tan Bl)
*/
double xin(bl, ra, xa)
double bl, ra, xa;
{
double tanbl = tan(bl);
double term1 = xa*(1 - tanbl*tanbl);
double term2 = (1 - ra*ra - xa*xa)* tanbl;
double term3 = (1 - xa*tanbl);
double term4 = (ra*tanbl);
return (term1 + term2) / ( term3*term3 + term4*term4) ;
}
/*
* 2 2
* (Ra - Ro) + Xa
* GAMMA = sqrt( ----------------- )
* 2 2
* (Ra + Ro) + Xa
*
* VSWR = (1 + GAMMA) / (1 - GAMMA)
*/
double swr(r0, ra, xa)
double r0, ra, xa;
{
double gamma = ((ra-r0)*(ra-r0) + xa*xa)/( (ra+r0)*(ra+r0) + xa*xa);
gamma = sqrt(gamma);
return (1.+gamma) / (1. - gamma);
}
doit(degrees)
double degrees;
{
double Ra, Xa;
Ra = 50 * rin( degrees * (M_PI/180.), 2., 0.);
Xa = 50 * xin( degrees * (M_PI/180.), 2., 0. );
printf("%3d degrees Ra %f Xa %f SWR %f\n",
(int)degrees, Ra, Xa, swr(50., Ra, Xa));
}
main() {
doit(33.);
doit(45.);
doit(80.);
doit(90.);
doit(150.);
doit(180.);
}
Kenneth Grimm
> In article <32jngh$s...@search01.news.aol.com>, Micron3 <mic...@aol.com> wrote:
>>
>> I am curious why the fact that the RF Analyst has a
>>microprocessor in it would affect the accuracy?
>
> Hi again, Terry. There are millions of digital bits roaming around
> inside a microcomputer. RF fields can change those bits. Somewhere
> inside the device they probably do an analog to digital conversion
> and a one bit error can be a 50 percent error. I would not trust any
> readings done with ordinary ham-grade test equipment in the presence
> of the antenna near field and especially not anything containing a
> microcomputer. Away from EM fields, it should operate just fine.
Maybe I'm dense, but what has this stuff about RF interference to
the microprocessor in the RF analyst have to do with the SWR readings
being given by the device? Doesn't the device include its own
generator?
>>What I was
>>trying to understand is the fact that I read a VSWR at the antenna
>>on 14.200 MHz of 4.3 - 1 and in the shack it reads 1.5 while on
>>21.200 I read 2,1 - 1 at the antenna and 2.4 - 1 in the shack?
>
> It's likely that the characteristic impedence at the antenna is not 50
> ohms and remember your SWR meter is standardized for 50 ohms. If
> the characteristic impedence at the point of measurement is not the
> same as your SWR meter, results may look wierd. For instance, in my
> earlier example of a 50 ohm antenna fed with one electrical wavelength
> of 450 ohm transmission line and an SWR of 9/1, a 50 ohm SWR meter at
> the generator will not read 9/1. You would need a 450 ohm SWR meter.
>>The gentleman at the store told me this is normal and I need to cut
>>the coax feed to 65 feet and then the VSWR curves would fall in
>>place as the VSWR varies up and down with length. Cheers, Terry KJ7F
>
> Maybe you could work this thing backwards. By measuring the SWR and
> impdeance at the generator end and knowing the length and matched-line
> loss of the transmission line, you can calculate the SWR and the
> impedance at the antenna. The adjustments on these multi-band verticals
> interact so that changing the SWR on one band will also change the SWR
> on the other bands. I would be tempted to feed it with ladder-line and
> an antenna tuner and forget the SWR.
Forget what the guy at the store told you.
Before you do anything else, you should check with the manufacturer.
Do they advertise a 50 ohm feed? If so, ask *them* how you are expected
to achieve a good match.
I have been able to get a satisfactory match to 50 ohm coax for every
commercial multiband vertical, cubical quad, and yagi that I have ever
bought. Likewise for those that I have built myself. I can't believe
that any reputable company would sell a multiband antenna without
providing for a 50 ohm match with a reasonable VSWR bandwidth.
The only time I have ever resorted to tuned feeders was when I was
building gain antennas made of wire...longwires, rhombics, curtains, etc.
Sure you can get a good match with ladder line and an antenna tuner,
but this, in my opinion, is usually an unnecessary complication for
the average ham...(whatever that is). :)
73,
/<en
___________________________________________________________
Kenneth D. Grimm K4XL
gr...@alison.sbc.edu
___________________________________________________________
Chris Scott
>
> Now...let's turn to the idea that changing the length of the transmission
> line would change the SWR...well yes of course, in your case...it is
>
> feed point impedance and the surge (characteristic) impedance of the
> transmission line will result in the *transmission line* acting
> as an impedance *transformer*...and by definition, the nature of the
> impedance transformation is *dependant* the *length* of the coax.
> This stuff is in the ARRL Handbook briefly and in more depth in the
> Antenna Mannual.
>
> Depending on the actual degree of mismatch *and* the absolute values
> involved, it is *not* unusual for SWR values to *soar* when the
> transmission line length is at the *sharply* climbing regions of the
> Tan(h) function of impedance transformation.
Me thinks you ought to look at the ole' Smith chart one more time...
I'm a broadcast engineer with experience tuning am-fm broadcast antennas
and networks...
First, aside from the (often small) *reduction* in VSWR along the line
due strictly to line *loss* (as a resistive pad would do) the *VSWR*
DOES NOT I repeat DOES NOT change along the line. Basic Smith chart
mechanics will prove this... remember the (constant) VSWR circle?
Even if you have say 25 ohms j 0 (non reactive) at the load end, with
a quarter wave section of 50 ohm line connected to the generator, the
VSWR is still 2:1 at both ends (neglecting the small effect of line
loss). The 100 ohm j 0 presented to the generator is 2:1 just as 25
ohms is 2:1, and this constant VSWR continues along the line. The only
thing that changes along the line is the ratio of reactance to
resistance, which repeat every half wave.
The Smith chart also lets you calculat the VSWR at the load end if you
know the VSWR at the generator end and the line loss. As long as you
have a well-designed system with line loss under, say 2-3 db, the amount
of VSWR "improvement" will be small.
NOW, understand that most inexpensive "load analysers" are basically
reflectometer-type devices dressed in sheeps clothing. This type of
instrument is quite crude for accurate measurements. A real network
analyser will run you $ 15 K ++ . I personally use a GR 1606A Impedance
bridge, and GR 1602B admittance meter, each of which can be had for
about $500 at a hamfest.
Good Luck
Chris, WB9NEQ
--
SCO...@WKUVX1.WKU.EDU aka Chris Scott- C/E Public Radio- Western KY U
Telco: (502) 745-3834 Hm & Fax: (502) 781-1232
...just another insignificant VAX user. \\\//
(o o)
--------------------------------------------ooO-(_)-Ooo-----------------------
Gary Coffman
[I've deleted much of what Steve said, because I agree with most of it.]
>Now...let's turn to the idea that changing the length of the transmission
>line would change the SWR...well yes of course, in your case...it is
>almost unimaginable that the feed point of your antenna is 50 ohms
>resistive on any but a single frequency...but in fact, is probably
>never anywhere near 50 ohms over most of its operation range...and
>who knows what the reactive component is....
>
>It follows that a mismatch between the antenna
>feed point impedance and the surge (characteristic) impedance of the
>transmission line will result in the *transmission line* acting
>as an impedance *transformer*...and by definition, the nature of the
>impedance transformation is *dependant* the *length* of the coax.
>This stuff is in the ARRL Handbook briefly and in more depth in the
>Antenna Mannual.
>
>Depending on the actual degree of mismatch *and* the absolute values
>involved, it is *not* unusual for SWR values to *soar* when the
>transmission line length is at the *sharply* climbing regions of the
>Tan(h) function of impedance transformation. Solution? Lengthen or
>shorten the transmission line...or put an appropriate open-or-shorted
>shut about 0.1 wavelength from the antenna feed point to make the
>transmission line go flat...solutions that are easy on a single band
>but may require considerable experimentation when frequency hopping
>multiple bands is involved (as is the case of your multi-band antenna).
I don't think you intended to leave this impression, but it seems
you are saying that changing coax length will change the SWR on
the coax. That's false (aside from the relatively minor loss issues
already addressed by others). A transmission line transformer will
present a different impedance at it's generator end depending on
length and frequency, but the SWR on the line is wholely dependent
on the impedance of the antenna and the characteristic impedance
of the line. For any given mismatch at the antenna feedpoint, at a
given frequency, changing the length of the line will *not* change
the SWR on that line. A look at the Smith chart shows that when
using a transmission line as a transmission line transformer, you
rotate around a *constant VSWR circle* to get the impedance transformation
you desire. The distance around that circle is the length of line
you use. The SWR is *the same* at every point along the line.
>Now let's visit what the factory rep was trying to get across to you.
>
>Why not measure at the antenna (feedpoint)?
>
>Because you are interacting with the antenna by being there...because
>if you are interested in taming SWR with a noise bridge, reading
>at the antenna will not tell (inexperienced) you very much that *you*
>know how to *use*....this is not an insult: your post makes it clear
>that you do not have enough understanding to use at-antenna readings
>to advantage....that might not be the case for Roy Lewellen or me,
>but you are the one struggling along...hence it is better to watch
>the SWR at the point where it plugs into your transmitter when
>adjusting the antenna.
More to the point, a noise bridge *does not* measure VSWR. It measures
the presented impedance at that point along the transmission line
transformer. You have to work backwards with the Smith chart to
determine the true mismatch at the load and calculate the VSWR from
that.
Even if you were to use an actual VSWR measuring instrument, such
as a reflectometer, you'd still have the currents *on the outside
of the coax shield* to contend with. These will upset the VSWR
instrument reading. The velocity factor on the *outside* of the
coax shield is near 1 while the velocity factor *inside* the coax
is typically near 0.66, so the out of phase situation will vary
*along the length* of the cable. In fact, a good test for RF on
the *outside* of a coax cable is to take two VSWR readings at
different distances along the cable. If they differ by more than
can be accounted for by coax loss, you have RF on the outside of
the shield. That's certainly the case here. The coax is acting as
part of the antenna radiating system.
Gary
--
Gary Coffman KE4ZV | You make it, | gatech!wa4mei!ke4zv!gary
Destructive Testing Systems | we break it. | uunet!rsiatl!ke4zv!gary
534 Shannon Way | Guaranteed! | emory!kd4nc!ke4zv!gary
Lawrenceville, GA 30244 | | ga...@ke4zv.atl.ga.us
Cecil_...@ccm.ch.intel.com
<Cecil_...@ccm.ch.intel.com> wrote:
>
>Your analyzer is probably affected by the near
>field of the antenna and will give bad readings especially if it has a
>microcomputer in it. (I'm a microcontroller jock).
Oops, Steve, AB4EL, pointed out the error of my ways here. I was assuming
that the ham transmitter had to be on to make this reading. I obviously
knew nothing about the Autek RF-1. Steve tells me that it has a built-in
RF generator and therefore very little near-field radiation, certainly not
enough to affect the microcomputer. My statement applies to any measurement
attempted near the antenna with the ham transmitter power on. My apologies
to Autek Research, 4143 W. Waters Ave, no. 120, Tampa, FL 33614. The RF-1
is described in Aug QST page 47 and looks like a great antenna test and
tuning tool. Wish I had designed it.
73, Cecil, KG7BK, OOTC (Not speaking for Intel)
Bob Levine
>Instead it had to be measured at the end of the feed line which had to be
>65 feet long and that
>the VSWR would change depending on the lenght of the feed line. I told
>him that I was under
>the impression that the VSWR on the feed line should not change with
>lenght unless the load,
>in this case the antenna were unbalanced. He launched into a long
>lecture about how this was
I had a similar conversation with Joe Reisart once regarding the
Cushcraft R7. All bands were as advertised except 20m, which
had a 2:1 SWR bandwidth of only about 150 khz. He suggested I
try changing the lenght of my feedline. After lots of trial
and error, when I added 6' to the run, my 2:1 SWR bandwidth
on 20m went to about 250khz and all other bands remained
virtually the same.
6 additional feet of RG213U certainly shouldn't change
feedline loss too much (feedline was already about 75')
so I never could figure it out either
---
------------------------------------------------------------
Bob Levine KD1GG 7J1AIS VK2GYN formerly KA1JFP
lev...@mc.com <--Internet email Phone(508) 256-1300 x247
kd...@wa1phy.ma <--Packet Mail FAX(508) 256-3599
------------------------------------------------------------
Zack Lau (KH6CP)
: Terry KJ7F
Much poorly formatted text deleted.
Please keep the line lengths short.
Feedline lengths can alter the SWR if there
are significant outside shield currents on
the coax. These currents are not cancelled
out by the currents on the center conductor.
Thus, by altering the length of the coax,
if the outside shield is part of the antenna,
you will change the antenna and thus the
feedpoint impedance, changing the SWR.
Wrapping the coax around a big ferrite
toroid can reduce these unwanted shield
currents.
--
Zack Lau KH6CP/1 2 way QRP WAS
8 States on 10 GHz
Internet: zl...@arrl.org 10 grids on 2304 MHz
Bill Turner
Organization: Eskimo North (206) For-Ever
Another factor to be aware of is current flow on the outside of
the coax caused by an unbalanced antenna. A vertical with no radials or
with very short radials is SEVERLY unbalanced and you will indeed have
such current. The coax itself then radiates and becomes in effect part
of the antenna. This explains why changing the length of the coax
affects SWR - you're actually changing part of the antenna. One way to
get rid of the effect would be to bury the coax a few inches and thereby
couple that energy into the earth. Lossy, and not really recommended,
but it would help. The better way is to install resonant radials.
Hope this helps.
73 de Bill, W7LZP
Stephen Modena
the SWR on a line remains constant from one end to the other. In your
world (of Smith Charts), it does! In the world of real ham radio operators,
it doesn't. Everyday hams often interchange "good impedance match"
with "low SWR"...what's the point in fighting it? :^) In practise, I've
seen SWR change with changing transmission line length...whether one
understands *why* it *appears* to be changing is another matter altogether.
The original poster is having problems...not the least of which is
"who to believe." He's getting advice that goes against his book
learning, but his book learning is not helping him in the face of
confusing readings from his Autek. I see this kind of confusion
all of the time...why does it continue? Transmission line theory
was well worked out 100 years ago, right? :^) Is the answer: the
Smith Chart and constant VSWR circles? I doubt it.
I'm fairly successful at antenna building--fixed, portable and and mobile.
My antennas seem to radiate effectively...my logbook is my testament.
Yet if there's a single piece of ham radio "equipment" that befuddles and
mystifies most hams, it's their antennas. This in itself is amazing,
given the "low parts count" of these devices. ;^)
I know how to use a Smith Chart, but I don't bother. I have a simple
noise bridge, but I use my MFJ SWR Analyser almost exclusively.
One day we were setting up for a Field Day. One group had built a
wire quad (according to my specs), and had spent hours trying to
get the "SWR to behave." Finally, I had time to look at their effort.
I looked the antenna construction over: it was fine. They showed me
how irratic the SWR was behaving. To "fix it," they had been changing
the size of the loops! I brought over my radio--with the
tuner and SWR Velcro'd to it, plugged in the antenna and VOILA!...it
worked great. Then I looked at the coax connectors on their
equipment, and found that the one on the SWR meter/tuner was so worn
that it wasn't making contact. That team had more than 100 years of
accumulated experince, yet somehow were convinced that the *antenna*
was at fault...rather than *themselves*.
I have about six or so "rules of thumb" for designing and adjusting
antenna-transmission line systems. Whenever I build or trouble shoot
an antenna system, I make sure all the solder joints are good;
that all connectors are good; that the VOM gives the right readings;
that I use a "known-to-be-good: rig/tuner/SWR meter" (mine, not yours);
any coax in the system is well choked at BOTH ends; feedline
segments have to be close to 1/2-wavelength. I do whatever it takes
to be sure that the system conforms to good layout practise so that it's
behavior *ought* to conform to theory.
Above all, I am able to formulate a systematic process of testing and
verifying *each* part of an antenna-transmission line system: I
prove the mechanical, electrical and RF integrity of each component
alone...and then as it is added to a growing train of parts. Even
if this requires disassembling everything, in the end it saves time
and frustration.
The original poster can not visualize what is happening with his
antenna system...he is confused, because he can not effectively
convert his book knowledge of antennas and transmission lines into
a plan to effectively trouble shoot his new antenna and it's
associated transmission line.
He is not alone. This is easily the commonest problems in
ham radio.
Smith invented his nomographic chart at the beginning of WWII in full
realization that the war would require more antennas and miles of
transmission lines to be built than existed at that point. He knew
that *most* field "engineers" would fail if they had to become
mathematical physicists in order to get all that stuff to work...
Smith knew that a trouble-shooters tool was needed...so he condensed
the physical laws of transmission lines into a chart that required
only a straightedge and a draftsman's compass to use. Smith was a
pretty smart guy.
The question is: are we advice givers and column writers being nearly
as "smart" in advising our fellow hams?
When I teach antennas and transmission lines to people studying for
their license, I begin by assuring them that *radio* works by magic.
I also tell them that most hams know squat about antennas and that
most of what they will be told about antennas is bullshit. If
they listen carefully, they will see that most hams will contradict
themselves within five minutes on the topic of "antennas." Their biggest
problem will be in deciding who knows what they are talking about...
and who doesn't. Who should they learn from...and who should they avoid?
I think the original poster is in that situation.
The original poster has one or more gross errors in his antenna-transmission
line set up...or the antenna was broken out-of-the-box. That encapsulates
the first decision point.
If he were to talk to the factory...the *factory*(!), they might give him
a suggestion or two on how to verify whether the antenna itself--alone,
is OK and properly tuned...and from that be able to decide whether this
problem should be solved by a warrantee replacement...or by ripping out
everything *except* the antenna and starting again. :^)
To the original poster: I recommend that you obtain and read:
"Baluns: What They Do and How They Do It"
Roy Lewallen, W7EL
Antenna Compendium: Volume 1; pp. 157-64
ARRL, Newington, 1985
ISBN 0-87259-019-4
In a nutshell, that article is likely to show you most of the practical
operational details that you will need to master to trouble shoot your
antenna problem. Don't let the title throw you: everything you really
need to *understand* is in that article. Roy is a pretty smart guy. :^)
73/Steve
arm...@delphi.com
Forrest Gehrke
length due to loss (which for small variations in length
will be tiny). Barring huge losses or high current on
the outside of the coax sheath, the SWR should be the
same at any point along the line. If the meter is
showing large variations in spite of that, the meter
has problems.
LE>I had a similar conversation with Joe Reisart once regarding the
LE>Cushcraft R7. All bands were as advertised except 20m, which
LE>had a 2:1 SWR bandwidth of only about 150 khz. He suggested I
LE>try changing the lenght of my feedline. After lots of trial
LE>and error, when I added 6' to the run, my 2:1 SWR bandwidth
LE>on 20m went to about 250khz and all other bands remained
LE>virtually the same.
LE>6 additional feet of RG213U certainly shouldn't change
LE>feedline loss too much (feedline was already about 75')
LE>so I never could figure it out either
I can't believe Joe told you that SWR will change with feeder
length. Rather I believe he suggested this because a change
in length will produce an input impedance which your
tuner or transmitter was better able to cope.
---
ž SLMR 2.1a ž * aibohphobia n. -- The fear of palindromes.
John F. Woods
>The original poster is having problems...not the least of which is
>"who to believe." He's getting advice that goes against his book
>learning, but his book learning is not helping him in the face of
>confusing readings from his Autek. I see this kind of confusion
>all of the time...why does it continue?
It continues because some people mistakenly assume that "book learning"
has nothing to do with the real world.
It is only in a fantasy world that SWR changes with line length, and
confusion continues because too many who know better indulge those who
don't and let them continue to believe that it does.
Yes, SWR readings on real-world meters often change when real-world
transmission line length is changed. But that is *never* because the
SWR changes, and indulging that fantasy that it does can *only* get in
the way of solving the problem; sometimes it fails to get in the way
enough to prevent the problem from being either solved or covered up,
and that lets the poor victim believe that, sure enough, you can change
the SWR by changing the transmission line length, leading to frustration
and wasted time in the future.
If, instead, you tell them that the bogus readings are caused by what
they're really caused by in the real world (busted meters, broken
connectors, RF on the shield, etc.) you point them in the right direction
not only to solve their current problem, but any future problems. Granted,
it is much harder to communicate this adequately than to just say "yeah,
dink around with the coax length until it works", but unless they've
bought real estate in Fantasy Land, it will have a lot more to do with
their antenna.
There's too much snake oil in the real world already. Don't manufacture
more of it.
By the way:
> feedline segments have to be close to 1/2-wavelength.
Why on Earth do you believe that? That means EXACTLY and ONLY that at
ONE particular frequency (and even harmonics thereof) the impedance of
the load is presented at the near end of the coax; in the real world,
the same magnitude of mismatch will be present regardless of the length.
(Granted, a fixed-tune rig might be happier with one kind of reactance
or another, in which case a PARTICULAR antenna will work better with
a half-wavelength line than a quarter-wavelength line, because that
PARTICULAR antenna had the right kind of reactive component -- but another
antenna will be at the worst possible reactance with a half-wavelength
feedline!)
You might as well apply leeches to your antenna, if you're not going to
bother understanding how it really works.
Gary Coffman
>I enjoyed being taken to task (nicely) by several of you about whether
>the SWR on a line remains constant from one end to the other. In your
>world (of Smith Charts), it does! In the world of real ham radio operators,
>it doesn't. Everyday hams often interchange "good impedance match"
>with "low SWR"...what's the point in fighting it? :^) In practise, I've
>seen SWR change with changing transmission line length...whether one
>understands *why* it *appears* to be changing is another matter altogether.
>
>The original poster is having problems...not the least of which is
>"who to believe." He's getting advice that goes against his book
>learning, but his book learning is not helping him in the face of
>confusing readings from his Autek. I see this kind of confusion
>all of the time...why does it continue? Transmission line theory
>was well worked out 100 years ago, right? :^) Is the answer: the
>Smith Chart and constant VSWR circles? I doubt it.
>The original poster can not visualize what is happening with his
>antenna system...he is confused, because he can not effectively
>convert his book knowledge of antennas and transmission lines into
>a plan to effectively trouble shoot his new antenna and it's
>associated transmission line.
>When I teach antennas and transmission lines to people studying for
>their license, I begin by assuring them that *radio* works by magic.
>I also tell them that most hams know squat about antennas and that
>most of what they will be told about antennas is bullshit.
Well then there's little wonder why they are confused. There's nothing
magic or occult about radio, transmission lines, or antennas. If fewer
people would treat them as *magic*, and bother to teach how they actually
work (by *applying* that book learning), there'd be lots fewer hams
who don't "know squat" about antennas and feed systems.
The very *first* thing amateurs should be taught is the difference
between SWR readings and impedance readings. That'd reduce confusion
markedly by itself. The "book learning" is *not* wrong. Antennas and
feedlines really do work the way the books say. SWR really *is* constant
along a transmission line. What's lacking is an understanding of measurement
technique, and how to interpret those readings for a given installation.
In this case, the factory "engineer" either didn't know what he was
talking about, or didn't know how to explain what he was talking
about. Or perhaps he *knew* the problem was one of a *design* that
makes the outside of the coax "hot" to RF but didn't want to admit
that to a customer so he fed the customer a line of BS instead.
You aren't helping by teaching misinformation and perpetuating the
myth that antennas and transmission lines work by "magic". They
work according to very rigorous laws of physics, and only rigorous
understanding of that can lead to a proper understanding of what
to do when things appear to go awry.
Bob Levine
-->
-->LE>6 additional feet of RG213U certainly shouldn't change
-->LE>feedline loss too much (feedline was already about 75')
-->LE>so I never could figure it out either
-->
-->I can't believe Joe told you that SWR will change with feeder
-->length. Rather I believe he suggested this because a change
-->in length will produce an input impedance which your
-->tuner or transmitter was better able to cope.
-->
-->---
--> ş SLMR 2.1a ş * aibohphobia n. -- The fear of palindromes.
My posting didnt say specifically the SWR, although close.
I he told me was that if I change the feedline length, the
2:1 swr bandwidth would change. It did. Anyhow, if you don't
believe what I posted, then ask him yourself.
Alternatively, put up an R7 and try it yourself.
K2BT de KD1GG
Tom Bruhns
: meter should only show variation with feedline
: length due to loss (which for small variations in length
: will be tiny). Barring huge losses or high current on
: the outside of the coax sheath, the SWR should be the
: same at any point along the line. If the meter is
: showing large variations in spite of that, the meter
: has problems.
One way this can happen: if your meter is not calibrated
for the line impedance you are measuring. For example,
let's say you have a 75 ohm line terminated in 50 ohms,
and you measure it with a meter calibrated for 50 ohms.
Then an integral number of 1/2 waves from the load, you
will measure close to 1:1 (assuming the loss isn't too
bad), but an odd number of 1/4 waves, you will measure
something in excess of 2:1 SWR (again assuming reasonably
low loss). With high loss on any load, or with a
matched load, you'd measure about 1.5:1 with this setup.
Moral: be sure your SWR meter is calibrated for the
impedance of the line you are measuring, if you want
meaningful measurements.
73, K7ITM
Micron3
respond with ideas and advice. I have been out of town on
vacation for several days and am amazed at the volume of
replies. There have been many well thought out replies, so
many in fact I have lost track of who said what. I have a few
quotes here and I hope the original posters will forgive me for
not taking the time to correctly identify them.
>The original poster can not visualize what is happening with his
>antenna system...he is confused, because he can not effectively
>convert his book knowledge of antennas and transmission lines into
>a plan to effectively trouble shoot his new antenna and it's
>associated transmission line.
Boy was this comment correct. This is my hobby and antennas
are not my career (thank goodness.)
>Even if you were to use an actual VSWR measuring instrument, such
>as a reflectometer, you'd still have the currents *on the outside
>of the coax shield* to contend with. These will upset the VSWR
>instrument reading. The velocity factor on the *outside* of the
>coax shield is near 1 while the velocity factor *inside* the coax
>is typically near 0.66, so the out of phase situation will vary
>*along the length* of the cable. In fact, a good test for RF on
>the *outside* of a coax cable is to take two VSWR readings at
>different distances along the cable. If they differ by more than
>can be accounted for by coax loss, you have RF on the outside of
>the shield. That's certainly the case here. The coax is acting as
>part of the antenna radiating system.
From reading everyones replies I think this best sums up the
situation. Obviously the antenna is very imbalanced and there
is current flowing on the outside of the cable. I think I am
hearing that this is messing up the readings I am getting, not
that the SWR is really changing. Could this be why the company
recommend 65 feet of feedline? The "Ham Quality" meter most
of us have will show a good match with that particular length?
>Failure to follow factory recommendations in trouble shooting
>means that your conversations with him are *probably* a waste
>of *his* time. The factory rep is 1,000 miles from your
>antenna...he may have a standardized procedure for analyzing
>problems *by telephone*...and I'd advise against rejecting his
>*requirements* in order to help you.
I haven't ruled out what the factory guy said. I was
trying to understand what he was trying to tell me rather
than waste his time on the phone. Also, I couldn't quite
bring myself to cutting up the feedline with out knowing
why I was doing it and that it was for a valid reason.
There was at least one comment on the fact that I
didn't mention by name the manufacturer of the antenna.
I did't mention it for several reasons. First, I didn't
want to come across as "flamming" the company for something
that very well might be my fault not theirs. Thats how
I would want to be treated if the shoe were on the other
foot. And second, I didn't want the brand of the antenna
to alter or color the responses.
The antenna is advertised as being a "No Tune" antenna
and the instruction book doesn't give any info on how to
change the band centers on any band other than 80 M
which requires changing a part that the factory supplies
upon request. I am using brand new RG-213 that I tested
with the connectors in place before installing it in the
system. It met the manufactures spec's for 100 feet of
that type of cable (at least with my Ham Quality test
equipment.)
I think it sounds like I should either choke off the
current flowing on the outside of the line or cut the
feedline to 65 feet and fool the SWR meter and rig into
thinking that the SWR is correct (which it may well be
if I could measure it with a broadcast quality meter.)
or fork out the green stuff for a high quality piece of test
equipment. I'll probably try the choke first if I can find the
stuff here in the wilds of Idaho since I don't have
to chop up the cable to do it and don't have the money
for the test equipment method.
Again thanks for the messages, I've learned a
great deal so far and I'm sure I'll learn more
before I'm finished.
Terry KJ7F
Stephen Modena
>
> From reading everyones replies I think this best sums up the
>situation. Obviously the antenna is very imbalanced and there
>is current flowing on the outside of the cable. I think I am
>hearing that this is messing up the readings I am getting, not
>...
>I was trying to understand what he was trying to tell me...
>.... Also, I couldn't quite
>bring myself to cutting up the feedline with out knowing
>why I was doing it and that it was for a valid reason.
>to alter or color the responses.
>
> The antenna is advertised as being a "No Tune" antenna
>and the instruction book doesn't give any info on how to
>change the band centers on any band other than 80 M
>which requires changing a part that the factory supplies
>upon request. I am using brand new RG-213 that I tested
>with the connectors in place before installing it in the
>
> I think it sounds like I should either choke off the
>current flowing on the outside of the line or cut the
>feedline to 65 feet and fool the SWR meter and rig into
>or fork out the green stuff for a high quality piece of test
>equipment. I'll probably try the choke first if I can find the
>stuff here in the wilds of Idaho since I don't have
>to chop up the cable to do it and don't have the money
>for the test equipment method.
>Terry KJ7F
Terry:
Here's my suggestion for choking the exterior current.
Buy an 8' length of *thin* PVC drain pipe...to be used for winding
your coax in a choking coil (at each end). After doing calculations,
cut two pieces of PVC to length needed.
Using the standard formula for the inductance of a solenoid:
L = ( D x D x N x N )/( [18 x D] + [40 x L] )
[where D=diameter in inches, L=length in inches, N=number of turns]
determine the number of turns needed to yield a reactance of 500 ohms
(10 times 50 ohms nominal), remembering that
Xl = 2 x Pi x F x L
[where Pi = 3.14, F=freq in Hz, L=inductance in Henries].
My calculations say (RG-213 winds 12 turns for 5 inches)....
wind 27 turns closed spaced, covering 11 inches, to achieve
an inductance of approx. 23 microHenries, yielding an RF
reactance of 506 ohms (i.e., 10 times nominal 50 ohms).
Check my math...excuse me, my arithmetic.
Those 27 turns will consume about 28-30 feet of coax...taped to the
PVC...leave a couple of feet before starting the
choke coil on the PVC....allowing easy connection to the antenna
and convenient lead away...and easy connection to your Autek
at the other end. Half the length of your coax will be used in
chokes (for testing)...and the other half allows you to try
different configurations of lead away from the antenna to
assure you are far enough away from the antenna .
Now you have 100 feet of uncut coax ready for measurements...
adequately choked on 80 meters, and very well choked on all
bands above (except if self resonance happens to fall in the
12 or 10 m bands).
If you get the same "funny" reading on each band with choking
as without...we are barking up the wrong tree. :^)
73, Steve
P.S. My recommendation of the standard solenoid formula is
based on the knowledge that it is close enough (+/- 10%) for
coils made with RG-58 and RG-8, as verified by resonating the
coax coil with a known capacitor and dip-metering it and cross
verifying it using a noise bridge and a 50 ohm series resistor.
Gary Coffman
> I think it sounds like I should either choke off the
>current flowing on the outside of the line or cut the
>feedline to 65 feet and fool the SWR meter and rig into
>thinking that the SWR is correct (which it may well be
>if I could measure it with a broadcast quality meter.)
>or fork out the green stuff for a high quality piece of test
>equipment. I'll probably try the choke first if I can find the
>stuff here in the wilds of Idaho since I don't have
>to chop up the cable to do it and don't have the money
>for the test equipment method.
You don't need better test equipment. You just need to understand
what the test equipment you have is *really* telling you. To a
very large extent, knowledge can substitute for money in antenna
work.
You have the stuff needed for a choke. Just wind several turns
of the coax in a tight loop (6 inch dia) near the antenna. This
will choke off RF currents on the shield. It's not as effective
at HF as a ferrite loaded coil, but it should make a noticeable
difference. If it *does* make a difference in your readings, but
not enough to totally solve the problem, you have two choices.
First you can *resonate* the coil, thus making it a parallel
trap, with a capacitor soldered to the shield across the coil.
That'll only work for a fairly narrow frequency range. Or second,
you can find some ferrite material to make the untuned choke more
effective. Amidon sells direct, and has suitable cores. Also study
the way you are routing the coax away from the antenna. To minimize
coupling and RF pickup on the coax shield, you want the coax to be
at right angles to the plane of the antenna elements.
A final thought. The antenna manufacturer may be *counting* on the
coax radiating to make the antenna work, IE it's really the coax
that makes up the bulk of the antenna on 80 meters. In that case,
first you've been *had*, and second you're going to have to follow
their advice to get it to work.
Chuck Hawley
> I think it sounds like I should either choke off the
>current flowing on the outside of the line or cut the
>feedline to 65 feet and fool the SWR meter and rig into
>thinking that the SWR is correct (which it may well be
>if I could measure it with a broadcast quality meter.)
>or fork out the green stuff for a high quality piece of test
>equipment. I'll probably try the choke first if I can find the
>stuff here in the wilds of Idaho since I don't have
>to chop up the cable to do it and don't have the money
>for the test equipment method.
Roll the first 35 feet of the coax (nearest the antenna) into abt
an 8 inch diameter close wound coil. Try not to get the ends next
to each other. Then see what you get. Air wound choke...no cutting.
No expense. You can do it in Idaho.
Chuck Hawley, KE9UW in Urbana, Illinois
haw...@aries.scs.uiuc.edu
School of Chemical Sciences, Electronic Services
University of Illinois, Urbana-Champaign
Micron3
(Gary Coffman) writes:
>A final thought. The antenna manufacturer may be *counting* on the
>coax radiating to make the antenna work, IE it's really the coax
>that makes up the bulk of the antenna on 80 meters. In that case,
>first you've been *had*, and second you're going to have to follow
>their advice to get it to work.
Gary,
That thought occured to me also after I sent the last message.
The antenna already has three 25 "Counterpoise" radials that
I understand affect the 40 M Band. They very well could be
counting on the feedline acting as part of the antenna in order
for it to work, especially on 80 M. If that is the case I suppose
I can either follow the directions and cut the feedline to 65 feet
or perhaps add some 65 foot counterpoise radials and see what
that does to things I don't mind following directions, I just wanted
to know the "Why" of it.
Terry KJ7F
Michael J Dower
ga...@ke4zv.atl.ga.us "Gary Coffman" writes:
> In article <32udiq$m...@search01.news.aol.com> mic...@aol.com (Micron3)
> writes:
> > I think it sounds like I should either choke off the
> >current flowing on the outside of the line or cut the
> >feedline to 65 feet and fool the SWR meter and rig into
> >thinking that the SWR is correct (which it may well be
> >if I could measure it with a broadcast quality meter.)
[snip]
> You have the stuff needed for a choke. Just wind several turns
> of the coax in a tight loop (6 inch dia) near the antenna. This
[snip]
>
> A final thought. The antenna manufacturer may be *counting* on the
> coax radiating to make the antenna work, IE it's really the coax
> that makes up the bulk of the antenna on 80 meters. In that case,
> first you've been *had*, and second you're going to have to follow
> their advice to get it to work.
Just a few questions / points ...
The above could be true, in which case perhaps the 65 feet of line needs to be
strung as a vertical radiator, with the "conventional" part of the antenna at
the top.
In that case, what would happen if the choke was placed on the feedline at 65
feet from the feedpoint? I think that the the line section between the choke
and transmitter would then behave as a line, while the section between the
choke and feedpoint would behave as a line *and* a radiating element. Perhaps
the line shield needs to be grounded at the 65 foot point, although if this is
a node for the shield's outside currents then it may not matter.
73's de VK2ENG
--
Michael J Dower
'Quoth the raven, "Never more".' ... Poe
Art Harris
In <32m4rk$p...@jabba.cybernetics.net>, AB...@jabba.cybernetics.net said:
SM>It follows that a mismatch between the antenna
SM>feed point impedance and the surge (characteristic) impedance of the
SM>transmission line will result in the *transmission line* acting
SM>as an impedance *transformer*...and by definition, the nature of the
SM>impedance transformation is *dependant* the *length* of the coax.
SM>This stuff is in the ARRL Handbook briefly and in more depth in the
SM>Antenna Mannual.
No argument there.
SM>Depending on the actual degree of mismatch *and* the absolute values
SM>involved, it is *not* unusual for SWR values to *soar* when the
SM>transmission line length is at the *sharply* climbing regions of the
SM>Tan(h) function of impedance transformation. Solution? Lengthen or
SM>shorten the transmission line.
It's not unusual for the impedance presented to the transmitter to vary
wildly with changing line lengths under mismatched conditions. However,
the VSWR will not change. VSWR is strictly a function of feedpoint
impedance and the characteristic impedance of the transmission line.
It is not unusual for VSWR *meter readings* to vary when the line length
is adjusted, but that is very different from saying the VSWR is really
changing.
Art N2AH
Art Harris
>In article <32tc64$g...@hopscotch.ksr.com>, John F. Woods <j...@ksr.com> wrote
>>It is only in a fantasy world that SWR changes with line length, and
>>confusion continues because too many who know better indulge those who
>>don't and let them continue to believe that it does.
>my 1990 ARRL Handbook, chapter 16, Fig. 24 says that 100 ft
>of coax with a matched-line loss of 6db (RG-58 on 2m) will show a VSWR
>of 1.5/1 at the input end while the VSWR at the load (antenna) end is
>10/1. If one cuts the line at 50 ft. one will measure a VSWR of 2.5/1.
>1.5/1 at one end, 2.5/1 in the middle, 10/1 at the load...
I believe both of the above statements are true!
The SWR does not change with line length. But the SWR *reading* at the
transmitter end of a lossy line will be lower than the true SWR. Again,
we're getting into an instrumentation problem. The fact that an
instrument is being used incorrectly or is giving incorrect readings
does not change the laws of physics.
The fellow who originally posed the question about SWR and line length
indicated that on some bands, he was actually reading a HIGHER SWR at
the transmitter end than at the antenna! Clearly this is an
instrumentation problem.
Art N2AH
Cecil_...@ccm.ch.intel.com
Art Harris <art.h...@woodybbs.com> wrote:
>The SWR does not change with line length. >Art N2AH
Hi Art, sounds like you mean well but please reference the definition of
SWR. The reflection coefficient is part of the definition of SWR and
is defined as the square root of the reflected power divided by the
forward power (measured at any point on a transmission line).
Assume a transmission line with 3db loss. Assume half the power at the
antenna is reflected. The reflection coefficient at the antenna will be
the square root of 1/2. The reflection coeffieicnt at the transmitter
will be the square root of 1/8. Clearly the SWR is different at the antenna
and the transmitter. If you transmit 8 watts, 4 watts reach the antenna,
2 watts are reflected at the antenna, and one watt of reflected power
reaches the transmitter end (assume that watt is not re-reflected).
If my 1+rho/1-rho math is correct the _actual_ SWR at the antenna is 5.8
and the _actual_ SWR at the transmitter is 2.1 and a properly functioning
SWR meter will yield those SWR readings. Add another equal length of
transmission line and the SWR measured at the transmitter will be 1.4
The SWR _does_ change with line length. If one doesn't believe it, one
should replace the 70 ft. of RG-58 on a G5RV with the same electrical
length of 9913... both 50 ohms. Especially on the higher frequency bands,
the SWR at the transmitter end will measurably increase, the antenna tuner
will work a little harder, and more power will be radiated. I know... that
was my first step in improving my G5RV.
Forrest Gehrke
CE>Hi Art, sounds like you mean well but please reference the definition of
Cecil, Can we drop this thread? A number of people, including
myself have pointed out that, yes, swr can change along
significantly different lengths of line for the frequency
being used, if the line is lossy.
However, the original poster of the question was wrestling with
a problem involving just a small difference in line length, as
I recall it, 6 feet. He had had others telling him that this
small change, resulting in being able to operate his transmitter
into that line whereas the original length he could not, was
due to SWR change.
Now you know as well as I that this is not true. In his case it
was simply the fact that the small difference in length was
resulting in an impedance his transmitter was more capable of
coping. This happens all the time, particularly with today's
power transistor outputs which have circuitry to reduce power
when faced with high SWR or impedances they can't deal with,
to prevent their destruction. Yes, this impedance transformation
is due to a real SWR, but not necessarily beyond the ability
of the transmitter to handle if within some acceptable range.
Assuming the SWR meter is calibrated for 50 ohms and we are
dealing with low loss 50 ohm transmission lines, and there is
no current on the outside of the sheath, we have all seen SWR
meters with do show significant SWR change along the line,
even with small length changes. The simple conclusion here is
that we have a defective SWR meter (more likely a defective
design).
BTW, this is not uncommon among the SWR meter products available
on the market, unfortunately. I have had my share of them.
--k2bt
---
ş SLMR 2.1a ş Somebody had to do it
Cecil Moore
Forrest Gehrke <forrest...@cencore.com> wrote:
>
>Cecil, Can we drop this thread? A number of people, including
Hi Forrest, Forgive me if I have somehow blundered. My newsreader
(TIN) says that your latest posting was the third in this thread,
that you had not posted to this thread before, that this thread
originated with a statement that SWR does not change with line
length (no particular length mentioned), that any differences in
SWR measurments at opposite ends of a transmission line are due to
"instrumentation problems", and I believe that is false.
73, Cecil, KG7BK, OOTC
Dana Myers
>Cecil_...@ccm.ch.intel.com writes:
>>In article <32tc64$g...@hopscotch.ksr.com>, John F. Woods <j...@ksr.com> wrote:
>>>It is only in a fantasy world that SWR changes with line length, and
>>>confusion continues because too many who know better indulge those who
>>>don't and let them continue to believe that it does.
>>of coax with a matched-line loss of 6db (RG-58 on 2m) will show a VSWR
>>of 1.5/1 at the input end while the VSWR at the load (antenna) end is
>>10/1.
>
>I know that coax losses reduce SWR by turning the reverse wave into heat.
>Now explain how decreasing the line length is supposed to cut the SWR, as
>has been proposed.
Not only does the line loss reduce the SWR by turning the reverse wave
into heat, it also reduces the size of the reverse wave by turning the
forward wave into heat. :-)
---
* Dana H. Myers KK6JQ, DoD#: j | Views expressed here are *
* (310) 348-6043 | mine and do not necessarily *
* Dana....@West.Sun.Com | reflect those of my employer *
* "Sir, over there.... is that a man?" *
Forrest Gehrke
MY>Newsgroup: rec.radio.amateur.antenna
MY>Organization: Sun Microsystem Computer Corporation
MY>Not only does the line loss reduce the SWR by turning the reverse wave
MY>into heat, it also reduces the size of the reverse wave by turning the
MY>forward wave into heat. :-)
Good point, that! The reason why loss calculations with transmission
lines make necessary hyperbolic transcendentals instead of circular.
C language has them in its standard library.
--k2bt
---
ş SLMR 2.1a ş She was only a CW Op's daughter, but she dididahdahdidit.