Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

ICOM 746 LOW POWER?

628 views
Skip to first unread message

David Krakowsky

unread,
Nov 30, 2001, 5:53:54 PM11/30/01
to
Hi group, I just bought a new Icom 746. Everything seems to work
normally except for the output power on two meters. I only show 60
watts into a dummy load. I know its supposed to put out 100 watts,
right? Am I doing something wrong or does it have to go back?
Thanks. Dave K2DHK

jcm

unread,
Nov 30, 2001, 7:41:37 PM11/30/01
to
send it back its a ongoing problem i had it back 2 times..
"David Krakowsky" <doth...@hotmail.com> wrote in message
news:141e2b2a.01113...@posting.google.com...

Ken Bjorklund

unread,
Nov 30, 2001, 9:04:12 PM11/30/01
to

Dave,

Make sure you are using a accurate VHF watt meter and good load. A resonant 2
meter antenna will suffice if you are not sure of the load. I have a 746 and
2100H. Using a inexpensive 144-220mhz MFJ 812B watt meter and a Ringo Ranger 2,
both radios provide the maximum specified RF power output at my shack ( 100 &
55 watts respectively ). In another reply to your posting, it was mentioned
that low power on 2 meters was a ongoing problem with the 746. I read these
newsgroups almost every day and that's the first I've heard of it. Just be sure
of your power readings before sending the radio back. Good luck.


73
Ken
KK4BH

M.E.

unread,
Nov 30, 2001, 9:24:44 PM11/30/01
to
Try a different meter.
If they both agree, THEN you might have a problem.

--
Respond to my "real" e-mail;
kk...@ev1.net


"David Krakowsky" <doth...@hotmail.com> wrote in message
news:141e2b2a.01113...@posting.google.com...

Gary - KJ6Q

unread,
Dec 1, 2001, 1:21:10 PM12/1/01
to
I agree - my 746, into a good match and with a good wattmeter shows nearly 110 watts out on 2
meters...

--
Gary - KJ6Q
**************
Two things we'll never see:
1. Perpetual motion...
2. A liberal Democrat complaining
about "Too much government"
"Ken Bjorklund" <kk4...@cs.comnospam> wrote in message
news:20011130210412...@mb-mp.news.cs.com...

Tim

unread,
Dec 1, 2001, 3:05:04 PM12/1/01
to
Many folks tend to make wattmeter cables that are a foot or two long.
Depending on the cable uised, these can end up close to a 1/4 wave on 2M,
and will act as a transformer, making that nice 50 Ohm dummy load look like
a much higher impedance.

Tim N6DRA

"David Krakowsky" <doth...@hotmail.com> wrote in message
news:141e2b2a.01113...@posting.google.com...

David Krakowsky

unread,
Dec 1, 2001, 11:58:27 PM12/1/01
to
It was the dummy load/watt meter, I substituted another VHF watt meter
and the 746 is putting out 100 watts on two.
Thanks for the suggestion.
Dave K2DHK


"Tim" <N6...@nc.rr.comnospam> wrote in message news:<QJaO7.118914$HA6.17...@typhoon.southeast.rr.com>...

Rich Griffiths

unread,
Dec 2, 2001, 10:30:21 AM12/2/01
to
If the load is indeed 50 Ohms and the coax is 50 Ohms characteristic
impedance, the length of the coax is irrelevant. A 1/4-wave line will
not make it look like a higher impedance.

... Rich

Rick

unread,
Dec 4, 2001, 5:34:48 PM12/4/01
to
On Sat, 01 Dec 2001 20:05:04 GMT, "Tim" <N6...@nc.rr.comnospam> wrote:

>Depending on the cable uised, these can end up close to a 1/4 wave on 2M,
>and will act as a transformer, making that nice 50 Ohm dummy load look like
>a much higher impedance.
>

Why would that be, if the cable used is a 50 ohm cable and the load is
50 ohms?

Rick

Sideband

unread,
Dec 4, 2001, 5:59:01 PM12/4/01
to
Because the impedance of the load is inverted at odd 1/4 wave intervals
in coax.

de AB8LR

-SSB

Ed Cregger

unread,
Dec 4, 2001, 8:39:36 PM12/4/01
to
A 1/4 wave length of transmission line inverts. A half wave
length repeats. Mystifying, isn't it? <G>

For real mind blowing info, study for the ship radar
endorsement for the GROL.

Ed Cregger, NM2K


"Rick " <rick...@hotmail.com> wrote in message
news:3c0d4f50...@news.eclipse.net...

Rich Griffiths

unread,
Dec 6, 2001, 6:14:23 PM12/6/01
to
Rick:

I'm amazed by the nonsense replies you got on this one. Those
responders apparently have a fuzzy understanding of transmission lines.

In fact, the answer to your question is that it would not be. A 50-Ohm
cable connected to a 50-Ohm load will deliver 50 Ohms at the opposite
end regardless of its length.

... Rich

Gary - KJ6Q

unread,
Dec 6, 2001, 10:41:54 PM12/6/01
to
UMMmmmm - then I guess things like phasing lines using coax to feed multi-element arrays were all a
big silly waste of time...

What a bunch of ignorant dopes those antenna engineers are...

--
Gary - KJ6Q
**************
Two things we'll never see:
1. Perpetual motion...
2. A liberal Democrat complaining
about "Too much government"

"Rich Griffiths" <ri...@one.net> wrote in message news:3C0FFBCF...@one.net...

Reg Edwards

unread,
Dec 7, 2001, 2:58:56 AM12/7/01
to
> A 50-Ohm cable connected to a 50-Ohm load
> will deliver 50 Ohms at the opposite
> end regardless of its length.
============================
Indeed it will. Unfortunately an antenna seldom has an impedance
of 50 ohms.

For example a 1/4-wave vertical has an impedance of around 37 ohms
and a 1/4-wave coax 50-ohm line transforms it up to 67 ohms.

At the centre of a 1/2-wave dipole we see 75 ohms and a 1/4-wave
50-ohm line transforms it down to 33 ohms.

If A and B are two impedances to be matched then a 1/4-wavelength
of line of impedance SqrRoot( A*B ) will do the job.

On the other hand transmitters and receivers (and their users)
seldom notice the relatively small departures from the 50-ohms they
work best into.
----
**************************
Regards, Reg, G4FGQ
Free Radio Design Software
http://www.g4fgq.com
**************************

Mark Keith

unread,
Dec 7, 2001, 1:51:58 PM12/7/01
to
"Gary - KJ6Q" <davi...@gte.net> wrote in message news:<6UWP7.655$K%4.14...@dfiatx1-snr1.gtei.net>...

> UMMmmmm - then I guess things like phasing lines using coax to feed multi-element arrays were all a
> big silly waste of time...
>
> What a bunch of ignorant dopes those antenna engineers are...
>
> --
> Gary - KJ6Q

But thats dealing with phasing. IE: adding an extra 1/4 wave to run
another element 90 degrees out of phase. They are referring to
impedance. There should be no change in impedance with changes in
length. 2 ft, 9 ft, 429 ft, shouldn't matter. The only thing that
should happen is a lower SWR as you get longer and longer due to the
losses starting to build up if you had a mismatch between the antenna
and feedline. If you note changes with length of coax, you would have
a common mode problem and have rf flowing on the shield. To make a
transformer, you would use a different value coax set up as a series
section, or as a stub. Having a 1/4 jumper of the same 50 ohm coax as
the feedline is just addding to the whole feedline length. It
shouldn't act as a transformer. MK

Rich Griffiths

unread,
Dec 7, 2001, 8:30:41 PM12/7/01
to
Reg, of course, you're also correct.

I thought, however, with the misunderstandings we're seeing here, it's
best to stick with a straight answer to Rick's question/observation:

IF the cable is 50 Ohms, and IF the load is 50 Ohms, then the length of
the cable is irrelevant.

As you suggest, if the "IF"s are way out of wack, then we can do
"magical" things by fiddling with the length of the line. :-)

73

... Rich

Rick

unread,
Dec 8, 2001, 10:02:06 AM12/8/01
to
On Thu, 06 Dec 2001 18:14:23 -0500, Rich Griffiths <ri...@one.net>
wrote:

A 50-Ohm
>cable connected to a 50-Ohm load will deliver 50 Ohms at the opposite
>end regardless of its length.


Of course. And you will notice the original poster said he was
attaching the cable to a dummy load (presumably 50 ohms).

Rick

Gary - KJ6Q

unread,
Dec 8, 2001, 1:00:27 PM12/8/01
to
THERE'S *theoretical* electronics - THEN there's *real-life*, PRACTICAL electronics!

We have seen a number of antenna types (bazookas, coax flattops, etc) using various lengths/types of
coax in their construction that are used to "magically" transform - or trick - the apparent
resonance and impedance of the system into something other than what it would ordinarily be
according to theory and materials and dimensions used.

The effect of demonstrated - or apparent - resonance and feedpoint impedance is easily demonstrated
on VHF/UHF, where the simple act of altering the coax feedline length by a mere foot or so has the
APPARENT effect of significantly altering the resonant frequency - and the IMPEDANCE along with
it - of the system in use. *I* did it as recently as last night on 2 meters, when installing a new
SWR bridge in line to the VHF output of my Icom 746 - the SWR went CONSIDERABLY higher as indicated
on both the bridge and the indicator on the 746 with the change I initially used in the
installation - yet NOTHING in the antenna itself was in any way changed. Trial and error with
different short lengths of interconnecting coax, as it has ALWAYS done thru the years, soon restored
APPARENT low SWR at the operating frequency I was wanting best match in the system. So what IS the
*actual, TRUE* resonant frequency of my 2 meter antenna at 50 ohms? I really don't KNOW - but I *am*
able to make my transmitter THINK it occurs at 144.200 simply by careful selection of FEEDLINE
LENGTH - and can as EASILY make the SAME antenna APPEAR resonant at 50 ohms at 146.500 - just as it
did when I first connected the new SWR bridge that changed the feedline length!


NOW, that ain't "scientific", or strictly according to "theoretical antenna engineering" - but IT IS
easily repeatable and demonstrable in REAL LIFE APPLICATIONS, and proves well enough to ME that the
APPARENT and USABLE impedance and resonance of an antenna system CAN be (artificially perhaps!) be
manipulated and altered by the simple juggling of coax length that SHOULDN'T (in theory) have ANY
effect upon either resonant frequency OR impedance of the system!

You may quote theoretical electronics to me all you wish - but *I know* what WORKS in actual
practice, regardless of technical bafflegab that might be used to explain it away or render it
irrelevent!

--
Gary - KJ6Q
**************
Two things we'll never see:
1. Perpetual motion...
2. A liberal Democrat complaining
about "Too much government"

"Mark Keith" <nm...@wt.net> wrote in message news:25eb70d7.01120...@posting.google.com...

Rich Griffiths

unread,
Dec 9, 2001, 12:35:18 PM12/9/01
to
Actually, what you describe *is* "scientific", or strictly according to
"theoretical antenna engineering". You apparently just don't understand
the science and engineering.

The starting point for this thread had to do with a 50-Ohm load on a
50-Ohm line. In that case, the line is "flat", and changing its length
will not change the impedance seen at the opposite end. In your
example, the antenna apparently is not a 50-Ohm load. So the line is
not flat, and changing the length will indeed change the impedance seen
at the opposite end.

It's all there in the theory that you sneer at. Get a good book, and
look up the transmission line equations.

... Rich W2RG

Gary - KJ6Q

unread,
Dec 9, 2001, 4:39:49 PM12/9/01
to
I don't "sneer" at engineering-theory - what I *DO* question is REAL LIFE APPLICATION of what we
hams ACTUALLY deal with on a regular basis! In the common employment of standard antennas -
especially pronounced on 2 meters and higher (what THIS thread is discussing...), virtually EVERY
antenna system *I* have built or bought has displayed EXACTLY the same characteristic of displaying
varying SWR depending on precise coaxial feedline length.

NOW, a ham relying strictly upon "engineering theory", would merely shrug his shoulders if his new 2
meter system loaded poorly due to high SWR, and ASSUME that NOTHING other than extensive redesign of
the antenna ITSELF (cursing that devious, incompetent manufacturer the whole time!) could correct
his "problem", when actually all he might need do is experiment with different feedline lengths in
order to obtain a properly functioning system... I mean, what POSSIBLE use could it DO to change
feedline length, when that "engineering Bible" sez it will DO NO GOOD, or have ANY effect in
altering resonance/impedence at the transmitter end? And how many commonly available beams DO
present a perfectly stable 50 ohm load at the desired operating frequency straight from the factory?

Blind acceptance of engineering fact/theory as an unyielding ABSOLUTE, without the ingenuity and
willingness to EXPERIMENT with usable and workable alternatives that seemingly are diametrically
OPPOSED to "known fact", place an artificial limitation that is totally foolish, arbitrary and
unnecessary.

You stick with your books, and *I* will continue to thrash on ignorantly with what has worked quite
well for ME in amateur radio for over 40 years...

--
Gary - KJ6Q
**************
Two things we'll never see:
1. Perpetual motion...
2. A liberal Democrat complaining
about "Too much government"

"Rich Griffiths" <ri...@one.net> wrote in message news:3C13A0D6...@one.net...

Dave Holford

unread,
Dec 9, 2001, 7:20:48 PM12/9/01
to

> Gary - KJ6Q wrote:
> >
> > The effect of demonstrated - or apparent - resonance and feedpoint impedance is easily demonstrated
> > on VHF/UHF, where the simple act of altering the coax feedline length by a mere foot or so has the
> > APPARENT effect of significantly altering the resonant frequency - and the IMPEDANCE along with
> > it - of the system in use.


The key word is 'system' - feedline and antenna.
You can do the same thing with a length of coax with no antenna
attached.

A chunk of co-ax, open at the far end will behave as a short circuit at
a quarter wavelength, and a high impedance (I avoided the temptation to
write infinite) at a half wave. Somewhere between 1/4 and 1/2 wave you
can find just about any impedance you like - of course the energy fed
into the system is dissipated in heating the co-ax but the transmitter
is happy.

Just did it to prove the point. Attached a half wave of open coax to the
transmitter and started to cut pieces off - soon found a point where the
transmitter happily provided maximum rated output - kept on cutting and
the power started to fall.

The transmitter was quite happy to feed full power into an open feedline
which, in complete accordance with theory, displayed an impedance of 50
ohms at the transmitter connector. BUT, only at that precise frequency.

With a real non-inductive 50 ohm load at the other end of the feedline I
would not expect any length of line, within reason, to have any effect
at the transmitter end.

Seems to me if you want the power to go to the antenna and not be wasted
in heating the feedline you should adjust the antenna instead of the
line.

Dave

Mark Keith

unread,
Dec 10, 2001, 1:05:50 AM12/10/01
to
"Gary - KJ6Q" <davi...@gte.net> wrote in message news:<%ysQ7.237$qS6....@dfiatx1-snr1.gtei.net>...

> THERE'S *theoretical* electronics - THEN there's *real-life*, PRACTICAL electronics!

They differ?


>
> We have seen a number of antenna types (bazookas, coax flattops, etc) using various lengths/types of
> coax in their construction that are used to "magically" transform - or trick - the apparent
> resonance and impedance of the system into something other than what it would ordinarily be
> according to theory and materials and dimensions used.

Maybe , but I usually prefer not to use such schemes unless I design
them that way in the first place. IE: using measured stubs, or
transformers of a different value coax.


>
> The effect of demonstrated - or apparent - resonance and feedpoint impedance is easily demonstrated
> on VHF/UHF, where the simple act of altering the coax feedline length by a mere foot or so has the
> APPARENT effect of significantly altering the resonant frequency - and the IMPEDANCE along with
> it - of the system in use.

It shouldn't though if everything is hunky dory. IE: the antenna
matches the feedline, and you don't have common mode currents flowing
on the shield.
If both of these are ok, it won't matter what the length is.

*I* did it as recently as last night on 2 meters, when installing a
new
> SWR bridge in line to the VHF output of my Icom 746 - the SWR went CONSIDERABLY higher as indicated
> on both the bridge and the indicator on the 746 with the change I initially used in the
> installation - yet NOTHING in the antenna itself was in any way changed.

Sure, you may a slight mismatch to the line, and you also may have
some common mode currents flowing on the shield.

Trial and error with
> different short lengths of interconnecting coax, as it has ALWAYS done thru the years, soon restored
> APPARENT low SWR at the operating frequency I was wanting best match in the system. So what IS the
> *actual, TRUE* resonant frequency of my 2 meter antenna at 50 ohms? I really don't KNOW - but I *am*
> able to make my transmitter THINK it occurs at 144.200 simply by careful selection of FEEDLINE
> LENGTH - and can as EASILY make the SAME antenna APPEAR resonant at 50 ohms at 146.500 - just as it
> did when I first connected the new SWR bridge that changed the feedline length!

I have no real gripe with you doing this, but it's not the method I
would take.
I would tune the antenna to the feedline, and make sure no common mode
currents reared their ugly heads. After that , it wouldn't matter what
length line I used. In all my coax fed antennas I have , not a single
one has ever been tuned by varying the feedline length. I have coax
fed dipoles and verticals on hf, and two yagi's , one for 6 and one
for 2 meters. None require me to tweak the feedline lengths. If I want
to move the resonance to another frequency, I have to get on the roof
and adjust the gamma match, say in the case of one of my yagi's. And
in that case, I couldn't move too far, unless I wanted to tweak
element lengths to keep the same gain/fb numbers.


>
>
> NOW, that ain't "scientific", or strictly according to "theoretical antenna engineering" - but IT IS
> easily repeatable and demonstrable in REAL LIFE APPLICATIONS, and proves well enough to ME that the
> APPARENT and USABLE impedance and resonance of an antenna system CAN be (artificially perhaps!) be
> manipulated and altered by the simple juggling of coax length that SHOULDN'T (in theory) have ANY
> effect upon either resonant frequency OR impedance of the system!

But in the scenario you describe, according to theory ,it does make
perfect sense. You have an apparant mismatch between your antenna and
the feedline. You also may have a small amount of common mode currents
flowing on the shield. In either or both of these cases, what you are
doing will work. It's just not my preferred route to go.


>
> You may quote theoretical electronics to me all you wish - but *I know* what WORKS in actual
> practice, regardless of technical bafflegab that might be used to explain it away or render it
> irrelevent!

Well, I don't think I've said anything that is overly technical. In
fact, I usually shy away from overly technical jabber. I *know* that
what you are doing is working for you. I'm just saying if I had to do
it, it would be done a different way. MK

Mark Keith

unread,
Dec 10, 2001, 1:31:41 AM12/10/01
to
"Gary - KJ6Q" <davi...@gte.net> wrote in message news:<FSQQ7.759$qS6.2...@dfiatx1-snr1.gtei.net>...

> I don't "sneer" at engineering-theory - what I *DO* question is REAL LIFE APPLICATION of what we
> hams ACTUALLY deal with on a regular basis! In the common employment of standard antennas -
> especially pronounced on 2 meters and higher (what THIS thread is discussing...), virtually EVERY
> antenna system *I* have built or bought has displayed EXACTLY the same characteristic of displaying
> varying SWR depending on precise coaxial feedline length.

I already covered that in another post... Mismatch, and common mode
currents...


>
> NOW, a ham relying strictly upon "engineering theory", would merely shrug his shoulders if his new 2
> meter system loaded poorly due to high SWR, and ASSUME that NOTHING other than extensive redesign of
> the antenna ITSELF (cursing that devious, incompetent manufacturer the whole time!) could correct
> his "problem", when actually all he might need do is experiment with different feedline lengths in
> order to obtain a properly functioning system...

No, I would correct the real problem. Either the antenna was not
correctly measured, or the matching device needed to be adjusted.

I mean, what POSSIBLE use could it DO to change
> feedline length, when that "engineering Bible" sez it will DO NO GOOD, or have ANY effect in
> altering resonance/impedence at the transmitter end? And how many commonly available beams DO
> present a perfectly stable 50 ohm load at the desired operating frequency straight from the factory?

Many. I know my was A4S right on the money the first time I built it.
I built a 3 el 6m yagi about 2 months ago, and again , no problems at
all. I had a decent match the first time I fired it up, and I was
using a gamma match. If I did have a mismatch, I would adjust the
gamma match, not the coax length.


>
> Blind acceptance of engineering fact/theory as an unyielding ABSOLUTE, without the ingenuity and
> willingness to EXPERIMENT with usable and workable alternatives that seemingly are diametrically
> OPPOSED to "known fact", place an artificial limitation that is totally foolish, arbitrary and
> unnecessary.


No, it's just that I've already tried what you do, mainly by accident,
and found out I like other methods better. If varying coax length
varies the SWR, something is not quite right with the antenna to
feedline connection/match or the antenna is poorly decoupled.


>
> You stick with your books, and *I* will continue to thrash on ignorantly with what has worked quite
> well for ME in amateur radio for over 40 years...

No problems with me...It's your antenna. MK

Gary - KJ6Q

unread,
Dec 10, 2001, 1:29:59 PM12/10/01
to
DAVE sez:

"With a real non-inductive 50 ohm load at the other end of the feedline I
would not expect any length of line, within reason, to have any effect
at the transmitter end"

Which HAS BEEN *precisely* MY point!

In actual, real life practice, there are PRECIOUS FEW "real, non-inductive 50 ohm load(s)) in actaul
use by hams - especially as we get into VHF/UHF frequencies!

SO, to grandly quote textbook "rules" as absolutes in such cases is both unrealistic and foolish!

The sinple act of installing coax connectors has the unavoidable effect of creating minute
alterations in feedline impedence - and by the time you include a SWR bridge, a coax switch, and
another 4 to 6 coax fittings to interconnect all this stuff together, what losses and impedence are
you NOW operating under?

THIS is the REAL WORLD many (most?) of us are actually operating in, and in such cases, smuggly
quoting meaningless textbook responses is just THAT, MEANINGLESS!

I don't challenge for a MOMENT the textbook, theoritical "truth" as it applies to feedlines - I
merely want to stress that very FEW of us are using the pure, unblemished installations that those
textbooks are designed to address... And to arbitrarily IGNORE that fact, and deny the readily
available work-arounds to restore our access to a reasonably properly operating system due to
artifical constraints imposed by blind faith in "textbook truth", is self defeating.

--
Gary - KJ6Q
**************
Two things we'll never see:
1. Perpetual motion...
2. A liberal Democrat complaining
about "Too much government"

"Dave Holford" <hol...@sympatico.ca> wrote in message news:3C13FFE0...@sympatico.ca...

Gary - KJ6Q

unread,
Dec 10, 2001, 2:26:19 PM12/10/01
to
"Mark Keith" <nm...@wt.net> wrote

Gary - KJ6Q said:

> > NOW, a ham relying strictly upon "engineering theory", would merely shrug his shoulders if his
new 2
> > meter system loaded poorly due to high SWR, and ASSUME that NOTHING other than extensive
redesign of
> > the antenna ITSELF (cursing that devious, incompetent manufacturer the whole time!) could
correct
> > his "problem", when actually all he might need do is experiment with different feedline lengths
in
> > order to obtain a properly functioning system...
>
> No, I would correct the real problem. Either the antenna was not
> correctly measured, or the matching device needed to be adjusted.
>
>

Which sorta brings us to the crux of the issue - assuming we both have a reasonably properly
designed identical antenna installations, and the simple act of installation has introduced
variables that has now created a VSWR of, say, 1 1/2 to 1 that we both find unacceptable, YOU would
spend several hours, or more, lowering your tower to effect a change of match at the ANTENNA
itself - while *I* spend several MINUTES doing the SAME THING (in practical effect and end result)
at the transmitter end with a simple swap in feedline length... AND in such a case, can you REALLY
and HONESTLY tell me that *anyone* on the distant end of our subsequent QSOs will in the SLIGHTEST
detect ANY difference in our relative signals?

I don't argue that YOUR way is incorrect - only that it is needlessly labor intensive, and the
difference in end result will likely be the same in BOTH cases - only YOURS came from a textbook,
and mine came from actual. practical experience. YES, *I* have done it BOTH ways, and found the end
results to be identical - so why beat myself up doing it the "textbook" way?

I certainly agree as to the basic importance of a good understanding as to the principles and
theories of antenna/feedline matching - just as it is to recognize the available work-arounds that
MAY provide a practical and effective alternative solution, essentially delivering the same result
with less effort - after all, we don'r ALL relish the notion of labor intensive antenna adjustments
in wind, rain or snow - especially when an added foot or two of feedline in the warmth of our
hamshack might be just as effective and lots faster!

And keep in mind I have been referring here to relatively MINOR mismatches - not radical or major
design or material flaws that I claim can or should then be corrected simply by feedline
manipulation! I maintain that up to a certain point, the SIMPLE and less "technically correct" way
will deliver essentially the SAME level of effectiveness and efficiency as the more "textbook
correct" method. Sloppy? Perhaps, in the eyes of a purist, but pretty effective and practical in
real life!

Rich Griffiths

unread,
Dec 10, 2001, 3:59:19 PM12/10/01
to
Phew! It's amazing how fast the postings fly when you leave computer
for a bit! Enough has gone back and forth that I won't beat the
technical issues anymore. Actually it seems that the various postings
don't differ all that much over the technical realities.

But you know, Gary, it does appear that you're sneering at the theory.
I think the problem faced by the ham who merely shrugs his shoulders if
his new 2 meter system loaded poorly would not be that he relied
strictly on engineering theory, but that he relied too little. Also, I
think the flaws you attribute to the literature ("engineering bible")
aren't there. My books have lots on the variation of impedance along a
line and the effects of stubs and other matching tools. And I know of
at least one ham (Cecil Moore) who has been quite outspoken about these
methods; his web site has (had?) a special section on "my feedline tunes
my antenna".

Do most hams know enough about this? I don't think so. Remember, the
starting point for this thread was that a few hams thought you could
change the impedance at the end of a flat line by changing its length.
Do most hams know enough about digital data transmission or microwave
equipment? I don't think so. And we probably never will. But the
solution to the issues you bring up is to continue learning more theory,
not less. Because the theory is not in error. It's *very* good.

... Rich

Mark Keith

unread,
Dec 11, 2001, 3:06:11 AM12/11/01
to
"Gary - KJ6Q" <davi...@gte.net> wrote in message news:<v%7R7.152$vo1....@dfiatx1-snr1.gtei.net>...

> "Mark Keith" <nm...@wt.net> wrote
>
> Gary - KJ6Q said:
>
> > > NOW, a ham relying strictly upon "engineering theory", would merely shrug his shoulders if his
> new 2
> > > meter system loaded poorly due to high SWR, and ASSUME that NOTHING other than extensive
> redesign of
> > > the antenna ITSELF (cursing that devious, incompetent manufacturer the whole time!) could
> correct
> > > his "problem", when actually all he might need do is experiment with different feedline lengths
> in
> > > order to obtain a properly functioning system...
> >
> > No, I would correct the real problem. Either the antenna was not
> > correctly measured, or the matching device needed to be adjusted.
> >
> >
>
> Which sorta brings us to the crux of the issue - assuming we both have a reasonably properly
> designed identical antenna installations, and the simple act of installation has introduced
> variables that has now created a VSWR of, say, 1 1/2 to 1 that we both find unacceptable,

More than likely I wouldn't find that unacceptable. 1.5:1 is fine with
me usually. Doing any tricks to improve that likely wouldn't help
much, unless you needed a wider usable bandwidth.

YOU would
> spend several hours, or more, lowering your tower to effect a change of match at the ANTENNA
> itself - while *I* spend several MINUTES doing the SAME THING (in practical effect and end result)
> at the transmitter end with a simple swap in feedline length... AND in such a case, can you REALLY
> and HONESTLY tell me that *anyone* on the distant end of our subsequent QSOs will in the SLIGHTEST
> detect ANY difference in our relative signals?

I don't think you will see any improvemant by adjusting the SWR if you
start out below 2:1. So in that area, we would both be wasting out
time. HF in particular. Now VHF/UHF is where I start to differ. If you
have an antenna on those bands that shows a large variation in SWR if
you add to the feedline, I'd be worried about common mode currents
more than anything. These currents will eat your lunch as far as
performance on those bands as they will skew the pattern up off the
horizon. So if the antenna in question showed this problem, and you
"fixed" it by adding line to get a better match, and I fixed mine by
adding better decoupling, my antenna would win. Hands down I bet.

>
> I don't argue that YOUR way is incorrect - only that it is needlessly labor intensive, and the
> difference in end result will likely be the same in BOTH cases - only YOURS came from a textbook,
> and mine came from actual. practical experience.

No, mine is from experience too. What makes you think I never do this
in the real world?

YES, *I* have done it BOTH ways, and found the end
> results to be identical - so why beat myself up doing it the "textbook" way?

Mainly common mode problems if that were the case.


>
> I certainly agree as to the basic importance of a good understanding as to the principles and
> theories of antenna/feedline matching - just as it is to recognize the available work-arounds that
> MAY provide a practical and effective alternative solution, essentially delivering the same result
> with less effort - after all, we don'r ALL relish the notion of labor intensive antenna adjustments
> in wind, rain or snow - especially when an added foot or two of feedline in the warmth of our
> hamshack might be just as effective and lots faster!

Oh, in general I'm fairly lazy really. I don't like to take my mast
down unless I have to. I had to do it yesterday, and it's still down.
Why did I take it down? I'm getting AM-BC spurs on 160m, and I'm
trying to find the source. Believe me, I had better things to do. But
the AM spurs are driving me crazy.
I thought I had a metal to metal "diode effect" in the area, but I
never did pin it down yesterday. My antennas are still down low as I
want to check more tomorrow.


>
> And keep in mind I have been referring here to relatively MINOR mismatches - not radical or major
> design or material flaws that I claim can or should then be corrected simply by feedline
> manipulation! I maintain that up to a certain point, the SIMPLE and less "technically correct" way
> will deliver essentially the SAME level of effectiveness and efficiency as the more "textbook
> correct" method. Sloppy? Perhaps, in the eyes of a purist, but pretty effective and practical in
> real life!

Many times yes, mainly on HF, but on VHF/UHF, I disagree. Common mode
currents are the usual suspect there. Or should be the first thing
looked into anyway.
And I'm well aware of Cecils setup, but he is using ladder line, not
coax. His setup was designed that way. MK

0 new messages