Q about balanced feed line
 

In article ,
Roy Lewallen wrote:

Big Endian wrote:

Here is something strange I discovered this weekend. While using a F.S.
meter as a RF sniffer, I was probing along some lamp cords that are
plugged into one of those extention outlet strips. I have a split block
ferrite core which measures one inch square and a 1/2 inch ID hole. So
while monitoring the FS meter I placed the core on the 115 to the house
outlet thinking that I will see a decrease in FS reading, much to my
surprise just the opposite happened, the FS meter pegged out. Why would
adding the core cause the FS reading to increase?

There are at least three possible explanations.

The first is one I've come across many times in doing EMI work, and is
probably the most likely. What happens is that you've got two or more
radiating sources whose fields cancel or partially cancel at the field
strength meter. When you reduce the radiation from one of those sources,
the field at the meter increases.

The second can be a bit subtle. Suppose you have a wire near an antenna
and that wire is, say, 3/4 wavelength long. Very little current will be
induced in this wire because it's far from self-resonance. Now put a
choke in the wire 1/4 wavelength from one end. Presto, a lot of induced
current in the now-isolated 1/2 wavelength portion. This phenomenon can
cause common mode current to increase when you add a common mode choke
on the feedline, if the current is being induced in the feedline (as
opposed to conducted) and the wire length and choke position are
favorable for this to happen.

The third is that the core you're using is a high frequency ferrite. If
it is, it will act as a loading coil, which could make a previously
non-resonant system resonant. I don't think this is likely, though,
because most clamp-on cores are made from ferrites suitable for EMI
suppression. A common type of material for this purpose is type 43
ferrite, which has a Q of 1 at a few MHz. So this type of ferrite won't
cause resonant effects like a high frequency ferrite (e.g., type 61) would.

Roy Lewallen, W7EL

I suppose the right thing to do is to place ferrite cores on all the
conductors in the area. Could get expensive. I have so many wires all
over the place the job seems a bit hopeless.

Thanks for the information Roy.

Q about balanced feed line
 

Big Endian wrote:

I suppose the right thing to do is to place ferrite cores on all the
conductors in the area. Could get expensive. I have so many wires all
over the place the job seems a bit hopeless.
. . .

Well, that's one way to solve the problem. Like the sign over my desk
says, "Anyone can design a bridge that will stand up. It takes an
engineer to design a bridge that will barely stand up."

Roy Lewallen, W7EL

Q about balanced feed line
 

Big Endian wrote:
In article ,
Roy Lewallen wrote:

Big Endian wrote:

Here is something strange I discovered this weekend. While using a F.S.
meter as a RF sniffer, I was probing along some lamp cords that are
plugged into one of those extention outlet strips. I have a split block
ferrite core which measures one inch square and a 1/2 inch ID hole. So
while monitoring the FS meter I placed the core on the 115 to the house
outlet thinking that I will see a decrease in FS reading, much to my
surprise just the opposite happened, the FS meter pegged out. Why would
adding the core cause the FS reading to increase?

There are at least three possible explanations.

The first is one I've come across many times in doing EMI work, and is
probably the most likely. What happens is that you've got two or more
radiating sources whose fields cancel or partially cancel at the field
strength meter. When you reduce the radiation from one of those sources,
the field at the meter increases.

The second can be a bit subtle. Suppose you have a wire near an antenna
and that wire is, say, 3/4 wavelength long. Very little current will be
induced in this wire because it's far from self-resonance. Now put a
choke in the wire 1/4 wavelength from one end. Presto, a lot of induced
current in the now-isolated 1/2 wavelength portion. This phenomenon can
cause common mode current to increase when you add a common mode choke
on the feedline, if the current is being induced in the feedline (as
opposed to conducted) and the wire length and choke position are
favorable for this to happen.

The third is that the core you're using is a high frequency ferrite. If
it is, it will act as a loading coil, which could make a previously
non-resonant system resonant. I don't think this is likely, though,
because most clamp-on cores are made from ferrites suitable for EMI
suppression. A common type of material for this purpose is type 43
ferrite, which has a Q of 1 at a few MHz. So this type of ferrite won't
cause resonant effects like a high frequency ferrite (e.g., type 61) would.

Roy Lewallen, W7EL

I suppose the right thing to do is to place ferrite cores on all the
conductors in the area. Could get expensive. I have so many wires all
over the place the job seems a bit hopeless.

Thanks for the information Roy.

Another option is to place a choke in the mains supply for the entire
station. If the RF current path is down the feedline (in common mode)
and then out into the mains, you can reduce it by inserting a choke at
any point along its path, and the mains supply may be more convenient
place.

The mains supply would need a different kind of RF choke, of course. One
option is to wind the entire mains cable - one or both live conductors,
neutral and safety ground - on a stack of large toroids (bearing in mind
what Roy said above, about choosing the right material). Another option
is to buy a ready-made mains filter that also has an RF choke in the
ground lead.

The difficulty is to organize the mains wiring so that *everything*
passes through the choke. If you leave even one sneak path to ground,
the RF will happily use it! Also any other lines such as phone and
network connections need RF chokes of their own. If this starts to read
like the things you have to do for lightning protection... why, that's
perfectly true because very similar considerations apply.

Groveling on your knees under the table is a chore, and of course it's
totally beneath the dignity of a Licensed Radio Armature. Well, do it
anyway. At my old QTH I found it well worth the effort. Re-grouping all
the mains feeds through a single filter cleaned up my mains-borne TVI,
and it also reduced the computer and other noise that was coming *up*
the mains.

Another very useful tool is a clip-on RF current meter. This will tell
you much more than a FS meter, because at last you can *see* where the
RF current is. For details, see my "Best of 'In Practice'" pages or the
MFJ catalog.



--
73 from Ian GM3SEK 'In Practice' columnist for RadCom (RSGB)
http://www.ifwtech.co.uk/g3sek

Q about balanced feed line
 

Ian White GM3SEK wrote:
Another very useful tool is a clip-on RF current meter. This will tell
you much more than a FS meter, because at last you can *see* where the
RF current is. For details, see my "Best of 'In Practice'" pages or the
MFJ catalog.

I just ordered an MFJ-853 and it's on 2-4 weeks backorder.
--
73, Cecil http://www.qsl.net/w5dxp

Q about balanced feed line
 

On Tue, 28 Feb 2006 04:00:37 -0800, Roy Lewallen
wrote:

Big Endian wrote:

I suppose the right thing to do is to place ferrite cores on all the
conductors in the area. Could get expensive. I have so many wires all
over the place the job seems a bit hopeless.
. . .

Well, that's one way to solve the problem. Like the sign over my desk
says, "Anyone can design a bridge that will stand up. It takes an
engineer to design a bridge that will barely stand up."

Or not [g]

http://www.ketchum.org/bridgecollapse.html

Q about balanced feed line
 

Cecil Moore wrote:
Ian White GM3SEK wrote:
Another very useful tool is a clip-on RF current meter. This will
tell you much more than a FS meter, because at last you can *see*
where the RF current is. For details, see my "Best of 'In Practice'"
pages or the MFJ catalog.

I just ordered an MFJ-853 and it's on 2-4 weeks backorder.

Same story over here - I'm waiting for samples to review for the
magazine.


--
73 from Ian GM3SEK 'In Practice' columnist for RadCom (RSGB)
http://www.ifwtech.co.uk/g3sek

Q about balanced feed line
 

Wes Stewart wrote:
On Tue, 28 Feb 2006 04:00:37 -0800, Roy Lewallen
wrote:

Big Endian wrote:
I suppose the right thing to do is to place ferrite cores on all the
conductors in the area. Could get expensive. I have so many wires all
over the place the job seems a bit hopeless.
. . .
Well, that's one way to solve the problem. Like the sign over my desk
says, "Anyone can design a bridge that will stand up. It takes an
engineer to design a bridge that will barely stand up."

Or not [g]

http://www.ketchum.org/bridgecollapse.html

Ah, yes, the Tacoma Narrows bridge. That movie played continuously in
the lobby of the Engineering Building at the U. of Colorado most of the
time I was going there. For anyone interested in this topic, I recommend
_To Engineer is Human: The Role of Failure in Successful Design_ by
Henry Petroski.

The whole trick, of course, is to stay on the right side of "barely".

Roy Lewallen, W7EL

Q about balanced feed line
 

Roy Lewallen wrote:
The whole trick, of course, is to stay on the right side of "barely".

Why not the left side? :-)
--
73, Cecil http://www.qsl.net/w5dxp

Q about balanced feed line
 

I keep forgetting some of these principles.

Thanks very much for the detailed and helpful explanations, Roy.

73,

Chuck

Roy Lewallen wrote:
chuck wrote:

As a follow-up, is there a practical way to determine how much current
unbalance will cause a one dB reduction in power delivered to the
antenna, the "lost power" being that power radiated by the
transmission line?

It seems like a rather complex modeling problem.

Thanks!

Chuck, NT3G


Nope. You can't generally say that one part of an antenna is radiating a
particular amount of the total power. Each part of the antenna creates a
field, and it interacts with the fields from all other parts of the
antenna. The total power radiated has to equal the total power input
less loss, but that's all you can say for sure. An example will help
illustrate the problem. Consider a parasitic element in a Yagi. It has
considerable current and contributes a great deal to the overall
pattern. Yet the total power input to the Yagi element is zero. With
zero power input, it can't, by itself, be radiating any power. What it
does is intercept some of the power radiated by the driven element and
re-radiates it with a different phase and amplitude. So how would you
apportion the power radiated by the driven element and the parasitic
element?

You might take a look at the current in the driven element and note that
it increases or decreases as you put the parasitic element in place and
remove it. But the current can either increase or decrease, depending on
the length and spacing of the parasitic element. So has the parasitic
element increased or decreased the power radiated by the driven element?
There's no answer.

You can look at the change in pattern in some idealized cases by
modeling. This can tell you what range of effects you might expect in a
real situation.

Roy Lewallen, W7EL