Ian White (GM3SEK), Cecil Moore (W5DXP), and probably others have
discused HF chokes made from flat vs. bunched coils of coaxial cable,
but I must have missed some posting(s), for I haven't recognized
anything that looks like a CONCLUSION.

* Is "bunched" or "flat" better?
* Are they basically single-band or wide-band or somewhere in-between?
* Are there "rules of thumb" to make them for hams with
* No test equipment?
* A grid-dip meter?
* A noise bridge?
* An antenna analyzer?

--
--Myron A. Calhoun.
Five boxes preserve our freedoms: soap, ballot, witness, jury, and cartridge
NRA Life Member and Rifle, Pistol, & Home Firearm Safety Certified Instructor
Certified Instructor for the Kansas Concealed-Carry Handgun license

On 24 Oct 2006 21:16:02 -0500, wrote:

I haven't recognized anything that looks like a CONCLUSION.

Hi Myron,

One conclusion is that there is an optimal solution, a special
solution, and a general solution, but it is more a problem of what was
the "ultimate question?"

In short, one size does not fit all and you need to define your
requirement to best obtain the right answer.

This is probably why the beaded ferrite choke is so popular. It might
not present 10KOhms of Z, but it doesn't have to, to be useful and it
is useful for a considerable stretch of spectrum.

73's
Richard Clark, KB7QHC

wrote:
Ian White (GM3SEK), Cecil Moore (W5DXP), and probably others have
discused HF chokes made from flat vs. bunched coils of coaxial cable,
but I must have missed some posting(s), for I haven't recognized
anything that looks like a CONCLUSION.

Ian and I agree that the ARRL was overly optimistic when
they asserted that a coiled coax choke can cover all three
octaves of HF.

The requirements for a choke feeding an
all-HF-band, ladder-line fed dipole are much higher than
the requirements for a choke feeding a resonant fan dipole.

* Is "bunched" or "flat" better?

Bunching disadvantage: Lowers the maximum choking impedance
and decreases the effective bandwidth. (Bunching increases
the stray capacitance.)

Bunching advantage?: Lowers the parallel self-resonant frequency
so not as many turns are required for narrow-band operation.

* Are they basically single-band or wide-band or somewhere in-between?

Depends upon the minimum required choking impedance. A 2:1 to
3:1 frequency range might be a good rule of thumb, e.g. one
choke can cover a tribander's 2:1 frequency range.

* Are there "rules of thumb" to make them for hams with
* No test equipment?
* A grid-dip meter?
* A noise bridge?
* An antenna analyzer?

How about: Using a 2L pop bottle as the coil form, don't use
more turns than the number of meters in a wavelength, e.g. no
more than 10 turns on 10 meters? That would probably work for
20m-6m as a rule of thumb.

The MFJ-259B will measure impedance up to 650 ohms. One could
define a "bandwidth" based on that 650 ohm value and knowledge
of a typical impedance waveform between those two points. For
instance, my 8 turn, 5.35" diameter choke falls below 650 ohms
at 4.63 MHz and 28.1 MHz. It might be useful over a 3:1 frequency
range from 40m-15m.
--
73, Cecil http://www.w5dxp.com

Cecil Moore wrote:
wrote:
Ian White (GM3SEK), Cecil Moore (W5DXP), and probably others have
discused HF chokes made from flat vs. bunched coils of coaxial cable,
but I must have missed some posting(s), for I haven't recognized
anything that looks like a CONCLUSION.


Sorry, my Internet access is down at the moment, so there's no knowing
when this will get out.


Ian and I agree

.... about quite a lot actually :-)


that the ARRL was overly optimistic when
they asserted that a coiled coax choke can cover all three
octaves of HF.

Agreed. That claim is made in the general ARRL Handbook. The ARRL
Antenna Book gives more detail, but notably does not include that claim.

The requirements for a choke feeding an
all-HF-band, ladder-line fed dipole are much higher than
the requirements for a choke feeding a resonant fan dipole.

* Is "bunched" or "flat" better?

"Bunched" and "flat" both mean the same to me - the cable is wound into
a flat coil like a length of rope. The actual choice is between that
shape and a cylindrical solenoid shape.


Bunching disadvantage: Lowers the maximum choking impedance
and decreases the effective bandwidth. (Bunching increases
the stray capacitance.)

Bunching advantage?: Lowers the parallel self-resonant frequency
so not as many turns are required for narrow-band operation.

The bunched/flat shape has more self-capacitance between turns, so for
the same total length of cable it has a lower self-resonant frequency.
The bandwidth of either shape is wide enough to cover any amateur band
with a very high impedance - way higher than you could get with a
ferrite choke - but you do need to get the resonant frequency right.



* Are they basically single-band or wide-band or somewhere in-between?

Depends upon the minimum required choking impedance.

A resonant coiled-cable choke is unbeatable on a single band, but you
may not actually *need* a choking impedance of several kohms.

If you only need an impedance of a few hundred ohms (though more is
always OK of course) then I'd agree with Cecil:
..
A 2:1 to
3:1 frequency range might be a good rule of thumb, e.g. one
choke can cover a tribander's 2:1 frequency range.

That seems realistic for either the bunched/flat or the solenoid shape.


* Are there "rules of thumb" to make them for hams with
* No test equipment?
* A grid-dip meter?
* A noise bridge?
* An antenna analyzer?

How about: Using a 2L pop bottle as the coil form, don't use
more turns than the number of meters in a wavelength, e.g. no
more than 10 turns on 10 meters? That would probably work for
20m-6m as a rule of thumb.

The MFJ-259B will measure impedance up to 650 ohms. One could
define a "bandwidth" based on that 650 ohm value and knowledge
of a typical impedance waveform between those two points. For
instance, my 8 turn, 5.35" diameter choke falls below 650 ohms
at 4.63 MHz and 28.1 MHz. It might be useful over a 3:1 frequency
range from 40m-15m.

You could certainly try Cecil's suggestions, though the resonant
frequency can be quite critical if you want to spread the performance
across more than one band.

A GDO should be OK for measuring the resonant frequency, with the two
ends of the choke shorted together.

However, maybe we're all trying to measure the wrong thing. The reason
why we're using these chokes is to reduce the unwanted common-mode
current, so why not cut to the chase, and measure the current itself?

Whoever you are, if you don't already have a snap-on RF current meter,
then make it your next homebrew project or ham radio purchase. It is THE
most useful single tool for RFI investigations, so it's always good to
have one in the shack.

In this particular application, you could measure the common-mode
current on the TX side of the choke - the answer is there "in a snap".
Then you can either declare the problem solved, or follow the meter
readings as you adjust the choke to reduce the current even further.

There are details of HB meters on my website, or the MFJ-854 is a good
commercial alternative (but definitely not the MFJ-805).



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

Ian White GM3SEK wrote:
Cecil Moore wrote:
Ian and I agree

... about quite a lot actually :-)

Lurkers may get the impression that we disagree a lot. :-)
Most of us technical types agree on 99 and 44/100% of
technical topics. We often argue about which favorite
approach is better when they are equally effective.

I just plotted the impedance of a lumped inductor in
parallel with a lumped capacitor, the impedance based
on a transmission line analysis, and the impedance
actually measured. The three graphs overlaid each other
until approaching the series resonant point where the
transmission line analysis more closely matched the
measured results. In particular, the actual measured
value was around +j258, the transmission line value was
around +j135, and the lumped circuit value was around
-j570.

"Bunched" and "flat" both mean the same to me - the cable is wound into
a flat coil like a length of rope. The actual choice is between that
shape and a cylindrical solenoid shape.

I don't think that's true, Ian. There are three ways to
wind a choke. 1. helical, 2. spiral, 3. random I'm pretty
sure "bunched" means "random" and *not* flat. Maybe that
is the origin of our disagreement over "bunched". If the
bunched coil was really random scatter winding, would you
like it as much?

There are details of HB meters on my website, or the MFJ-854 is a good
commercial alternative (but definitely not the MFJ-805).

I've got the MFJ-853. It snaps but doesn't unsnap. :-(
--
73, Cecil http://www.w5dxp.com

Cecil Moore wrote:
"Bunched" and "flat" both mean the same to me - the cable is wound
into a flat coil like a length of rope. The actual choice is between
that shape and a cylindrical solenoid shape.

I don't think that's true, Ian. There are three ways to
wind a choke. 1. helical, 2. spiral, 3. random I'm pretty
sure "bunched" means "random" and *not* flat. Maybe that
is the origin of our disagreement over "bunched". If the
bunched coil was really random scatter winding, would you
like it as much?

Sorry, maybe there's language barrier here. Imagine coiling a rope in
your hand like a cowboy, then laying the coil on the ground. I meant a
"flat" coil in that sense, as opposed to the other way of winding these
chokes on a cylindrical former.


There are details of HB meters on my website, or the MFJ-854 is a
good commercial alternative (but definitely not the MFJ-805).

I've got the MFJ-853. It snaps but doesn't unsnap. :-(

The 854 that I reviewed for Radcom was the same. We both got one that
had the core assembled upside-down, so instead of showing you how to
snap the plastic catch open, the instruction manual shows you how to
break it.



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

Ian White GM3SEK wrote:
Sorry, maybe there's language barrier here. Imagine coiling a rope in
your hand like a cowboy, then laying the coil on the ground. I meant a
"flat" coil in that sense, as opposed to the other way of winding these
chokes on a cylindrical former.

A lot of arguments are over semantics. Again, there are
three ways of winding the chokes; helical, flat spiral,
and bunched like a cowboy rope. The flat spiral is also
known as a "pancake" choke or coil.
--
73, Cecil http://www.w5dxp.com

Ian White GM3SEK wrote:
The 854 that I reviewed for Radcom was the same. We both got one that
had the core assembled upside-down, so instead of showing you how to
snap the plastic catch open, the instruction manual shows you how to
break it.

I'm trying to figure out how I can make more effective
measurements on coiled coax chokes with the equipment
that I presently have. I can measure the impedance with
my MFJ-259B if it is less than 650 ohms.

I also have a GDO. If I get an open circuit dip, would
that be the parallel resonant frequency? If I get a
short circuit dip, would that be the series resonant
frequency?

I also have a dual-trace 100 MHz Leader o'scope. I
should be able to find the resonant frequencies
by watching the voltage drop across a series resistor.
I also have the X-Y feature so I could look at the
source voltage, voltage across the choke, and voltage
across the resistor as the frequency is changed.

What do you think?
--
73, Cecil, http://www.qsl.net/w5dxp

On Wed, 25 Oct 2006 16:13:59 +0100, Ian White GM3SEK
wrote:

that the ARRL was overly optimistic when
they asserted that a coiled coax choke can cover
all three octaves of HF.

Agreed.

Hi Ian,

Really? 3 to 30 MHz is more than three octaves:
3 - 6 - 12 - 24 + 1/3 octave more

Are we talking about Ham Bands?
160 - 80 - 40 - 20 - 10

Are we restricting from the top of the 80M band through 10M?
4 - 8 - 16 - 32 (actually 88% octave more)

From 60M up?
60 - 30 - 15 + 75% octave more

A subset of HF? such as
160 - 80 - 40
80 - 40 - 20
40 - 20 - 10

To anyone,

Just where did the ARRL come up with this new band plan for HF?

Hi Myron,

Can you see any conclusion yet? Even with the new 3 Octave
specification you still have to pick your application from 3 choices.

73's
Richard Clark, KB7QHC

Richard Clark wrote:
Ian White GM3SEK wrote:
that the ARRL was overly optimistic when
they asserted that a coiled coax choke can cover
all three octaves of HF.

Agreed.

Really? 3 to 30 MHz is more than three octaves:
3 - 6 - 12 - 24 + 1/3 octave more

3 is not in the ham bands. 3.6 - 7.2 - 14.4 - 28.8
There's the three octaves the ARRL was talking about.
--
73, Cecil http://www.w5dxp.com