Balun with high SWR
Dave
Does anyone know where I can find information on the impedance
transformation that occurs when a balun is operated in the
presence of high SWR? I note that the ARRL Handbook is silent on
this point, simply stating that "loads with reactive components
should use appropriate networks to cancel the reactance." So I
take it that a 4:1 balun does not just divide the twin lead
impedances by 4.
I am specifically interested in feeding a non-resonant doublet
with twin lead, at QRP power, and want to get as much as possible
out the antenna and not burned up in the feedline or balun.
EZNEC gives a good idea of what the antenna impedance will be,
and TLW takes that to the output of the transmatch, but I can't
find any info on transforming that impedance through the balun to
the actual matching network.
Thanks,
Dave
AB9CA
W6RCecilA
> Does anyone know where I can find information on the impedance
> transformation that occurs when a balun is operated in the
> presence of high SWR?
Jerry Sevick has some information in his book, "Transmission Line
Transformers". The sad fact is that there is no balun that will
do the job you need done. Where are you going to get a balun
that will balance the currents when it is looking into a 3k+j3k ohm
load? My advice is not to allow the balun to see an impedance
far away from its design impedance. Baluns are not magic.
--
http://www.mindspring.com/~w6rca
-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 80,000 Newsgroups - 16 Different Servers! =-----
JR
ferrite will absorb all the rf. My first person experience of
flaming baluns I found it out the hard way. I did some
measurements afterwards as well with the burned up and
new baluns from various sources.
Once you get to 3-4X the normal output Z, the input seems to go
down to 5-6 ohms and losses zoom up.
fyi
Jerry AA2T
Tom W8JI
>Hi,
>
>Does anyone know where I can find information on the impedance
>transformation that occurs when a balun is operated in the
>presence of high SWR? I note that the ARRL Handbook is silent on
>this point, simply stating that "loads with reactive components
>should use appropriate networks to cancel the reactance." So I
>take it that a 4:1 balun does not just divide the twin lead
>impedances by 4.
Sevek has a lot of things right in his book, but his concept of what a
balun should do is wrong. The voltage baluns he favors are impedance
critical, and do not provide what you really want in most antenna
systems. They tend to be lossy and power limited except over a limted
range of impedances and frequencies.
If you are going to have a reactive load, or a mismatched load, stay
away from a voltage-type balun. In general, you also do not want a
balun with a step-up ratio when feeding balanced feeders on multiple
bands.
Choke baluns are virtually impedance non-critical, and generally are
1:1 ratio. They look like a very short section of transmission line
for normal-mode currents, so they do not modify the impedance much.
They look like a high impedance for common-mode currents that cause
problems.
The best choice is a choke balun, with a design selected for the power
levels you run. Stay away from strings of beads, because the choking
impedance is generally not high enough.
How much power are you going to run, and what type of antenna system?
I use a stack of four 61 material 2 inch diameter cores wound with 20
turns of twisted-pair HV teflon wire. That balun works very well from
1.8 to 30 MHz, and handles well over 1500 watts without great concern
about the feed impedance.
73 Tom
Dave
Thanks for the input.
>Where are you going to get a balun
>>that will balance the currents when it is looking into a 3k+j3k ohm
>>load?
Based on what you say and the comment in the Handbook, I take it
that transformers - which is all baluns really are - don't like
it when the current and voltage in the windings are out of phase.
Of course, the current and voltage are forced out of phase by the
transformer, but that is its own "natural" phase difference. The
high SWR forces other phase differences that the transformer
cannot accomodate?
I guess another concern is that with a high SWR there is power
being forced both dirctions simultaneously, i.e. both to and from
the antenna?
So what happens in the event of the 3k+j3k that you mention?
Does the loss within the balun rise sharply? Do we find that
instead of the 1 dB or less of loss typical of a balun, when the
high SWR is present that the loss goes up to 9 - 12 dB or higher?
I have seen several comments where hams have said that
such-and-such a balun burned up, caught on fire, or whatever,
implying a lot of power dissipated in the balun.
Dave
AB9CA
Dave
Hi Jerry,
Thanks for the info.
>>Once you get to 3-4X the normal output Z, the input seems to go
>>down to 5-6 ohms and losses zoom up.
An interesting comment. Are you saying that even if the core is
not saturated in the usual sense, with the conflicting currents
and voltages present, the core ceases to behave normally and
becomes basically a power sink rather than a power transfer
device?
When you ran your tests did you run them at low power levels so
the cores were no where near saturation.
Dave
AB9CA
Dave
Hi Tom,
>If you are going to have a reactive load, or a mismatched load, stay
>away from a voltage-type balun.
That is a comment I have seen other places. Unfortunately the
folks who make most of the antenna tuners appear to ignore that
advise. I know that the balun in my MFJ 949E is the voltage
variety.
>In general, you also do not want a
>balun with a step-up ratio when feeding balanced feeders on multiple
>bands.
Since none of the low-loss parallel-conductor feeder cable
happens to be 50 ohms, the only solutions would involve some
feeder arrangement the avoids a balun entirely? Either an honest
to gosh balanced-line transmatch, with a balun between the xmit
and the transmatch; or a length of impedance matching feeder.
What other ways are there to eliminate the balun (or high SWR on
the balun)? These are the only two I know of.
>How much power are you going to run, and what type of antenna system?
Power level is QRP (I'm trying to conserve milliwatts here!) and
the antenna will be something portable. Either a doublet or
vertical. They each have their pros and cons.
On a related thought, did I not see a comment on another thread
about your having looked into the effects of running parallel
conductor cable on or slightly under the ground? My plans are to
throw the twin lead feeder on the ground to run to the antenna
(about 50'). I don't envision that this will have much effect on
the feeder losses. But, since the ground has more loss than air,
but lots less than metal, I do expect it will unbalance the line
somewhat and result a small amount of line radiation.
Dave
AB9CA
On Fri, 11 May 2001 23:01:45 GMT, 2w...@contesting.com (Tom W8JI)
wrote:
Tom W8JI
>folks who make most of the antenna tuners appear to ignore that
>advise. I know that the balun in my MFJ 949E is the voltage
>variety.
Yes, while most tuners have current baluns the 949E has a voltage
balun.
>Since none of the low-loss parallel-conductor feeder cable
>happens to be 50 ohms, the only solutions would involve some
>feeder arrangement the avoids a balun entirely? Either an honest
>to gosh balanced-line transmatch, with a balun between the xmit
>and the transmatch; or a length of impedance matching feeder.
>What other ways are there to eliminate the balun (or high SWR on
>the balun)? These are the only two I know of.
There is an incorrect but common assumtion the balun ratio has
something to do with feedline design impedance. The impedance at the
end of a 450 ohm line feeding a typical multiband dipole is almost
never 450 ohms. It varies from a few ohms to many thousands of ohms.
Why would anyone care if the balun had a transformation ratio? The
impedance is all over the place! What you want is a wide-range balun,
and that is a 1:1 choking-type balun.
Second point, moving the balun to the input of the tuner does nothing
to improve balance of the system. If the tuner is a grounded center
true symmetrical tuner it can help balance, but most tuners are not
built that way. If they are constructed as a symmetrical grounded
center matching system, they are voltage-type balanced systems...and
that is undesirable anyway.
I leave my baluns on the output, because it is cheaper and better that
way.
>>How much power are you going to run, and what type of antenna system?
>
>Power level is QRP (I'm trying to conserve milliwatts here!) and
>the antenna will be something portable. Either a doublet or
>vertical. They each have their pros and cons.
Ahhh, then you task is more simple. You just need to scale down the
balun size I use while maintaning a high impedance. The balun in your
tuner can be rewired as a 1:1 choke balun, but I'd stack a few more
cores and use a higher voltage wire.
>On a related thought, did I not see a comment on another thread
>about your having looked into the effects of running parallel
>conductor cable on or slightly under the ground? My plans are to
>throw the twin lead feeder on the ground to run to the antenna
>(about 50'). I don't envision that this will have much effect on
>the feeder losses.
I measured some a year or so ago. The soil around the feeder just
killed the feedline, making it very high loss. That was true even
though I fished the feedline through large plastic (PVC) conduit.
I don't have the numbers now, but it was pretty bad.
73 Tom
W6RCecilA
> So what happens in the event of the 3k+j3k that you mention?
I don't know exactly what happens but the balun ceases to perform
the balun function.
There is an easy solution. Using 450 ohm ladder-line, the SWR for
3k+j3k is about 13:1. From that 3k+j3k point on the ladder-line,
if you back up 0.238WL, the impedance will be 34.6 ohm resistive.
And if you add 0.262WL, the impedance will be 34.6 ohm resistive.
This point is the current maximum point on the transmission line
and it repeats every half wavelength.
A 1:1 choke designed for 50 ohms will work reasonably well looking
into 34.6 ohms. I vary the length of my ladder-line until my choke
is operating at the current maximum point which is purely resistive.
Take a look at my All-HF-Band-Antenna (dipole) on my web page to
see how I do it.
john griffen
<non...@aol.com> writes
I use a 67 ft doublet with 18ft of 600 ohm open wire line. A 1:1 balun
is connected to the QRP rig and a balanced L network matches the
impedance at the Tx end of the feeder to 50 ohms.
Estimate the antenna impedance at the operating freq., calculate the Z
at the input end of the feeder then calculate values of L and C in the
network.
You will need to adjust L & C values a little to optimise, but the
networks I use for 80 and 40 m. give a perfect match for CW operation.
Reckon that the network losses are lower than those for a T network.
It works well - try it!
73s
John G3EGQ Cornwall
--
john griffen
Dave
Hi again Jerry,
>Once you get to 3-4X the normal output Z, the input seems to go
>down to 5-6 ohms and losses zoom up.
I have a follow up question on this. When you mention 3-4X the
normal Z, do you mean the magnitude of the line input impedance
or is this the value of R only? In other words if the input
impedance was, say 8-j198, the R would be only 8, but the
magnitude of the impedance would be about 198@-88deg, although it
is almost totally reactive impedance.
Would this example be about = to Z or would it be 1/40 of Z
(assuming a 4:1 balun fed by 50 ohms so Z=200)?
73,
Dave
AB9CA
Dave
Hi Tom,
>Why would anyone care if the balun had a transformation ratio? The
>impedance is all over the place!
One of the reasons for posting this question is that I have been
trying to calculate the L and C values needed for an antenna
tuner. To do this the impedance seen by the tuning network must
be known. I can get a rough idea of the impedance at the input
to the feedline through EZNEC and TLW. But the balun is between
the feedline and the network, and the balun is going to do
*something* to this impedance. If the impedance has little
reactance, the balun will - I presume - just divide it by the
impedance ratio, either 1, 4, or whatever. But if the reactance
is high? Well, that is my question.
>Second point, moving the balun to the input of the tuner does nothing
>to improve balance of the system.
Doesn't this depend on whether or not the transmatch is designed
to accept unbalanced feed? If it is designed to accept only
balanced feed than the unbalanced output of the xmit needs to be
isolated from ground, doesn't it?
>I measured some a year or so ago. The soil around the feeder just
>killed the feedline, making it very high loss. That was true even
>though I fished the feedline through large plastic (PVC) conduit.
>
This is not good news. If the balanced feedline loss goes way up
when lying on the ground, it makes the whole exercise moot.
73
Dave
AB9CA
On Sat, 12 May 2001 09:58:51 GMT, 2w...@contesting.com (Tom W8JI)
wrote:
>>That is a comment I have seen other places. Unfortunately the
Dave
Hi Cecil,
>Take a look at my All-HF-Band-Antenna (dipole) on my web page to
>see how I do it.
Without knowing that it was your's, I have already done that.
That is a very interesting antenna arrangement. Looks like it
ought to work fine.
After downloading and studying your page I ran some of the
antennas I am pondering through TLW looking for the low impedance
points. They are there all right. At 7 MHz it was about 58'
out, and at 14 MHz it was 43', and so on. I guess that so long
as the reactance is near zero and the resistance is between 25
and 100, the SWR seen by the coax ought to be less than 2:1.
Most xmiters will work OK with that.
But Tom mentions that lying twin lead on the ground - which is
what I was planning on doing since these are portable antennas -
will make the loss go way up. How much "way up" is, I'm not sure,
but it doesn't sound good. I would have to keep the feedline off
the ground or go to a resonant coax-fed antenna. Running a
portable vert and keeping the feedline off the ground is pretty
tough. Oh, well, something more to ponder. . .
Maybe its back to a long wire slung in the trees.
Thanks for your thoughts and time.
73,
Dave
AB9CA
Dave
Hi John,
Thanks for the input.
Sounds like your arrangement ought to work pretty well. Solves
just about all the problems. And it gets the balun out of any
high SWR. That may be what I end up doing.
I'm surprized that you can get 80M to work with such a short
wire. I get an input Z of 11-j1050 for 3.6 MHz. Looks like the
efficiency is not the best, with the input R so low, but it is
certainly better than no antenna at all.
73,
Dave
AB9CA
W6RCecilA
> This is not good news. If the balanced feedline loss goes way up
> when lying on the ground, it makes the whole exercise moot.
The good news is that it doesn't have to be very far off the ground.
The other night I was running A/B comparisons with my dipole and my
new vertical. The dipole won as expected. The next morning, I discovered
that the wind had blown my ladder-line off the cup hooks I use beneath
the eave of my house, and the ladder-line was laying on the grass. The
grass needed mowing so the ladder-line was a few inches off the ground.
Moral is: The fields around ladder-line are pretty well confined to
the space surrounding the ladder-line. A plastic fork every ten feet,
or something equally inventive, would probably be all that is needed.
Peter O. Brackett
Yes, I too have used antennas fed with 450 Ohm ladder line lying on the
"dry" ground, and they seemed to work fine.
I wouldn't leave it there for a long time, and I wouldn't do it if it was
raining or snowing or.... but on a nice sunny dry day, I didn't notice a
thing.
And if there is no danger of folks tripping on it, why then your suggestion
of holding it off the ground a few inches with some "plastic" picnic
"silverware" should work just fine.
I love messing with 450 Ohm and open wire ladder line, it's soo high tech...
and easy to use.
-Peter K1PO
"W6RCecilA" <Cecil....@IEEE.org> wrote in message
news:3B01A9E5...@IEEE.org...
john griffen
<non...@aol.com> writes
Here are the design details that were used......
67 ft doublet, assumed feedpoint Z 20 - j 1200 ohms.
At end of 18 ft of 611 ohm open wire line impedance calculates to 6.7 -
j 492 ohms (freq. 3.560 MHz, QRP CW chan).
One needs to match this Z to 50 + j 0 ohms.
For a bal. L network, series L in each leg needs to cancel the - j 246
ohms then one matches 6.7 + j 0 ohms to 50 + j 0 ohms using extra L and
a shunt C.
A shunt C value of 2250 pF was indicated. made up of a fixed and a large
value preset capacitor.
Adjustment of turns spacing on the coils and of C gave a perfect match.
I had previously found that the MFJ tuner with its 4:1 balun and T match
with this antenna was very inefficient at this freq. The L match works
well and ant. performance agrees with that expected from a shortened
doublet, just a few dB down on a full halfwave.
Hope this helps.
73 s
John Griffen G3EGQ
Cornwall
--
john griffen
Dave
will repost.
--------------------------------
Hi Tom,
>Why would anyone care if the balun had a transformation ratio? The
>impedance is all over the place!
One of the reasons for posting this question is that I have been
trying to calculate the L and C values needed for an antenna
tuner. To do this the impedance seen by the tuning network must
be known. I can get a rough idea of the impedance at the input
to the feedline through EZNEC and TLW. But the balun is between
the feedline and the network, and the balun is going to do
*something* to this impedance. If the impedance has little
reactance, the balun will - I presume - just divide it by the
impedance ratio, either 1, 4, or whatever. But if the reactance
is high? Well, that is my question.
>Second point, moving the balun to the input of the tuner does nothing
>to improve balance of the system.
Doesn't this depend on whether or not the transmatch is designed
to accept unbalanced feed? If it is designed to accept only
balanced feed than the unbalanced output of the xmit needs to be
isolated from ground, doesn't it?
>I measured some a year or so ago. The soil around the feeder just
>killed the feedline, making it very high loss. That was true even
>though I fished the feedline through large plastic (PVC) conduit.
>
This is not good news. If the balanced feedline loss goes way up
when lying on the ground, it makes the whole exercise moot.
73
Dave
AB9CA
On Sat, 12 May 2001 09:58:51 GMT, 2w...@contesting.com (Tom W8JI)
wrote:
>>That is a comment I have seen other places. Unfortunately the
Dave
will repost.
Hi Cecil,
>Take a look at my All-HF-Band-Antenna (dipole) on my web page to
>see how I do it.
Without knowing that it was your's, I have already done that.
That is a very interesting antenna arrangement. Looks like it
ought to work fine.
After downloading and studying your page I ran some of the
antennas I am pondering through TLW looking for the low impedance
points. They are there all right. At 7 MHz it was about 58'
out, and at 14 MHz it was 43', and so on. I guess that so long
as the reactance is near zero and the resistance is between 25
and 100, the SWR seen by the coax ought to be less than 2:1.
Most xmiters will work OK with that.
But Tom mentions that lying twin lead on the ground - which is
what I was planning on doing since these are portable antennas -
will make the loss go way up. How much "way up" is, I'm not sure,
but it doesn't sound good. I would have to keep the feedline off
the ground or go to a resonant coax-fed antenna. Running a
portable vert and keeping the feedline off the ground is pretty
tough. Oh, well, something more to ponder. . .
Maybe its back to a long wire slung in the trees.
Thanks for your thoughts and time.
73,
Dave
AB9CA