David <k4zzr@bellsouth.net> wrote:
>...I already have several of the FT-240-43 toroids and wonder how
>many turns of RG-213 coax and how tightly they need to be?
The winding on a toroid should be a single layer, not overlapping,
and it should be tight; i.e., the length of cable in the winding
should be as short as possible. You want to minimize parasitic
capacitance, to keep the self-resonant frequency of the choke high.
You want to operate below this self-resonant frequency. It helps a
little to break the winding into two halves on opposite sides of the
toroid, to keep the opposite ends of the winding far apart. (Be sure
to wind the two halves the same way, so that their H-fields add and
don't cancel!) I learned this trick from the RSGB's EMC book.
In my earlier response I was assuming smaller diameter coax. RG-213
is so thick that (1) there may not be room for 12 passes of it
through an FT-240 toroid; (2) _even_if_ a winding of RG-213 had the
same length, its capacitance would be greater than if you'd used
RG-58 or equivalent size; and (3) RG-213 can't be wound tightly, so
the winding must be longer, which increases its capacitance.
It may be more practical with thick coax to string a large number of
suitably-sized ferrite "beads" on it. But see below regarding how
large a number is required!
Consider using silver-plated & Teflon coax of less thickness for the
choke(s), if such coax can handle your transmitted power. I don't
know how much, but it's certainly better than ordinary copper and
polyethylene cable. Silver-Teflon coax is more expensive but you
won't need much if all you use it for is chokes. Of course, there'd
be connectors....
You can always avoid self-resonance trouble by making chokes of
lesser inductance and putting more of them in series along the line.
(See below regarding how many.) The limit is single-turn
(once-through) "bead" chokes, as mentioned above; however using
multiple turns makes much more effective use of the ferrite, which
costs more than the coax.
The inductance of a choke is proportional to the square of the number
of turns in its winding, and simply proportional to the number of
toroids in the core that the winding encircles. To limit capacitance
you must limit the number of turns in a winding, but you can get more
inductance by using more toroids in the core; and using more toroids
in the core does not require proportionally increasing the length of
the winding. If you use (say) eight toroids, rather than stacking
them in a single column, put them in two side-by side columns of four
toroids each, like the tubes of a pair of binoculars; and wind the
choke by passing the coax "up" through one column/tube and back
"down" through the other.
Whenever I've experimented with making ferrite-core chokes, I've used
an impedance meter such as the Autek Research RX Vector Analyst Model
VA1, which is advertised in QST for about $200. It's easy to
calculate the inductance of some number of turns on a core, by using
the core manufacturer's data sheet; but it's hard to predict the
_capacitance_ of a winding, or the self-resonant frequency -- which
is why you need a meter. I don't believe that a "dip" meter would be
useful in this application, because the Q of a ferrite choke is
typically very low. I imagine that an antenna noise bridge would be
useful for measuring the impedances of chokes, though -- just
remember to treat the choke as a two-terminal device; don't connect
the coaxial winding of the choke to the coaxial "antenna" connector
on the bridge, or you'll be looking through the choke in differential
rather than common mode.
Here's a sample set of impedance measurements I made, of one
FT-240-43 toroid wound with 5 turns of #18 wire:
| Impedance
Freq.| Mag. | Angle
(MHz)|(ohms)| (deg)
-----+------+--------
3.5 | 332 | not measured
14.0 | 235 | 8
28.0 | 203 | 18
The small angles mean that the impedance is mainly resistive, which
is good as long as there's not too much RF current flowing (so not
much power is dissipated; the dissipation is in the ferrite, and
heating the ferrite above its Curie temperature causes its
ferromagnetism to vanish, and your cable to melt). There won't be
much RF current if the chokes are doing their job. However, I
recommended an air-core choke at the antenna end of the cable because
with high power it's easy to overheat a ferrite toroid there.
The characteristic impedance, Zo, of the (accidental) transmission
line formed between the _outside_ of a coaxial line and other nearby
conductors (including the earth) is typically a few hundred ohms,
depending on spacing. To be at all effective, a choke or series of
chokes must have an impedance many times greater than Zo. Ten times
greater would provide around 20 dB of blocking, which may be
insufficient. Suppose Zo = 300 ohms. You won't get a choking
impedance of 10 times 300, or 3000 ohms, from one ferrite choke.
That's why you need more, in series.
The impedance of a single ferrite "bead" choke at HF depends on the
parameters of the bead, of course; but a ballpark value is 5 to 10
ohms. You'd need 300 to 600 beads to get 3000 ohms, and even 3000
ohms may not be sufficient.
In case you're thinking that no one would ever string 300 beads on a
coaxial cable, think again. Not long ago I read a technical article
by the well-known expert radio scientist and ham Jack Belrose VE2CV,
about a balanced antenna center-fed with coax, in which he described
stringing 300 beads on the coax. IIRC, he complained that 300 was
not enough.
Most commercial baluns have sufficient common-mode choking impedance
to get a reasonably symmetrical distribution of current on your
antenna, but not enough to attenuate common-mode current on your
feedline effectively for RFI/TVI prevention.
.
>I have a SPG right outside my shack and wonder if I should put the
>chokes outside *before* the surge suppressors?
Yes, because (1) you don't want much RF current flowing from your
antenna into your Single-Point-so-called-Ground, because it's not the
perfect current-sink that the term "ground" implies; and (2) you
don't want to insert impedance between your radio and your SPG.
>I have 4 TV's 19"-32" that get no TVI at all....but I'm getting RFI on my
>home entertainment surround sound system on the 57" large screen TV.
>Nothing on the screen, just donald duck on the audio. It happens both
>barefoot and a lot worse with 1.5kw. I've added brute force toroids
>to most of the lines to/from the receiver-amplifier but it's not
>working.
You say "most" of the lines. Maybe the rest of the lines need choking, too.
Maybe your chokes have insufficient impedance. See above.
>Now I wonder if the receiver is just being overloaded internally.
"Overload" suggests low-level input stages; but audio amps have
feedback, so it's often necessary to choke the speaker lines as well
as the power line and low-level cables.
Speaking of the power line, the AC power circuitry in a house is not
at all balanced for RF, so there's plenty of mode conversion between
common and difference modes. Therefore it's necessary to use
brute-force LC power-line filters that block both common and
difference mode RF current. Corcom makes good ones, and you can pick
them up cheaply at hamfests.
>Have you heard of any good surround sound systems less prone to RFI?
No, but I'm ignorant. My sound system is simple two-channel stereo.
It's completely free of RFI, even without chokes on its speaker
lines, because the chokes on my coax keep RF from getting back into
my house.
>Or any other suggestions you might add?
I fear I've written too much already. Sorry.
>Thanks for your help,
>David, K4ZZR
De nada.
73 de Chuck, W1HIS
> > From: Chuck Counselman <ccc@space.mit.edu>
To: <towertalk@contesting.com>
>> Date: Wed, 25 Sep 2002 13:04:58 -0400
>> To: "Carlos Augusto S. Pereira" <py4ro@superig.com.br>
> > Cc: <towertalk@contesting.com>, <rfi@contesting.com>
>> Subject: Re: [Towertalk] TV Cable Information and TVI
>>
>> At 8:18 PM -0300 9/24/02, Carlos Augusto S. Pereira wrote:
>>> ...the only place I have to bring the coax and rotor cables from the
>>> roof to the second floor (where is located my apartment) is the same
>>> place where the 75 ohms antenna tv cable passes. Does anyone have
>>> experienced the same situation? TVI will happen for sure?
>>
>> 1. It is important to put good common-mode-current RF chokes on both
>> your coax and your rotor cables. First, right at your Yagi feedpoint
>> put an air-wound choke coil in your coax, by winding six turns of the
> > coax closely, in a single layer (not bunched or overlapped), on a
>> 4-inch-diameter form such as PVC drain pipe. This choke will have
>> low power-dissipation but will not have enough choking impedance by
>> itself. Then you should have two good ferrite-core chokes, spaced at
>> intervals of about 2.5 meters (a quarter-wave at 10 m). A good
>> ferrite choke can be made by winding about 12 turns of your cable on
>> an Amidon type FT-240-43 or FT-240-77 toroidal core. ("FT" means
>> ferrite toroid. "240" means outside diameter = 2.4 inches, inside
>> diameter = 2.0 inches, and thickness about 0.4 inches IIRC. "43"
>> means type 43 ferrite, which has mu/mu0 of about 850 and is fairly
>> lossy at HF. Palomar is another source of such toroids. They're
>> made by Fair-Rite. I'm sure there are other sources outside the US.)
>> A toroidal-core choke like this has an inductance of around 35
>> microH, IIRC. Ferrite "beads" slipped or snapped over a cable are
>> _not_ good enough at HF. They don't have nearly enough impedance.
>> The chokes on your coax should be co-located with the chokes on your
>> rotor cable, so that one cable cannot carry RF current past the choke
>> on the other cable.
>>
>> 2. In addition, put good ferrite common-mode-current chokes on the
>> TV-antenna cable(s). If the TV cable(s) run(s) parallel to your
>> ham-antenna cable, then the common-mode-current chokes on the
>> TV-antenna cable(s) should be co-located with the chokes on your
>> rotor and ham-coax cables, so that one cable cannot carry RF current
>> past the choke on another cable.
>>
>> 3. If a TV antenna is near your Yagi, you should also insert a
>> high-pass filter in its feedline, to keep HF RF out of the
>> preamp/distribution amplifier.
>>
>> 5. If the TV preamp or distribution amplifier on your roof has a
>> separate cable for power, then put a ferrite choke on this pawer
>> cable, too. Same for the IF and power cables of any satellite-TV
>> downconverter.
>>
>> 6. Put a good low-pass filter at the output of your HF TX or xcvr.
>>
>> If you do all these things you should have no trouble with TVI. I
>> transmit 1.5 kW on all HF bands, 80 through 10 meters (except 30 m);
>> the coaxial feedline of my ham antenna runs right by my TV antenna
>> and very close to its coaxial feedline, yet I have no TVI on any
>> channel. In my case I also inserted a second high-pass filter at the
>> downstairs end of the TV cable, and I also inserted common-mode
>> chokes at the downstairs ends of all cables. This may have been
>> overkill, but chokes are cheap and you may as well do all of them at
>> once.
>>
>> An important other benefit of common-mode chokes is that they stop
>> RFI from the electric power wiring inside your building, and from the
>> sweep circuits of the TV sets in your building, from being conducted
>> up to your ham antenna. This makes a huge difference on the
>> lower-frequency ham bands.
>>
> > 73 de Chuck, W1HIS
>
|