Toroids frequency range and baluns

[James Duffey]

Bob - You wrote:

"Why are choke baluns wound as a coil vastly superior to choke baluns  
made with a sting of ferrite beads? With the coil type, you're adding  
additional coax loss... What am I missing here?"

I didn't see an explicit reply to this on the list, but the short  
answer is that with a coil balun, the inductance increases as the  
square of the number of turns. With the beads, the inductance just  
increases as the number of the beads. So, ignoring stray capacitance,  
a toroid with 10 turns will have 10 times the inductance of the same  
cable passed through 10 toroids (beads). You get more bang for your  
buck by coiling the cable. The additional losses are not too great for  
most applications. - Duffey
--
KK6MC
James Duffey
Cedar Crest NM





_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Jim Brown-10]

On Sat, 27 Sep 2008 09:21:11 -0600, James Duffey wrote:

>"Why are choke baluns wound as a coil vastly superior to choke baluns  
>made with a sting of ferrite beads? With the coil type, you're adding  
>additional coax loss... What am I missing here?"

>I didn't see an explicit reply to this on the list,

In the interest of brevity on the list, I referred those interested in
this to the tutorial I wrote on RFI and the use of Ferrite Chokes. That
material is anything but brief, but I believe that it is quite clear
and easy to study IF you understand the fundamentals of electrical
circuits that include R, L, C, and Z.

>but the short  
>answer is that with a coil balun, the inductance increases as the  
>square of the number of turns. With the beads, the inductance just  
>increases as the number of the beads.

YES

>So, ignoring stray capacitance,

You CANNOT ignore stray capacitance, it makes a VERY major contribution
to the behavior of ANY choke.

>a toroid with 10 turns will have 10 times the inductance of the same  
>cable passed through 10 toroids (beads). You get more bang for your  
>buck by coiling the cable. The additional losses are not too great for  
>most applications.

YES.

BUT -- it is NOT about INDUCTANCE, it is about the RESISTIVE component
of the impedance. And that RESISTANCE is the loss component of the
parallel RLC circuit formed by the inductance of the choke, the
capacitance, and the loss coupled from the ferrite core. It is the
RESISTANCE that solves our problems, NOT the inductance. We use
inductance in TRANSFORMERS and in resonant circuits that are part of
radios. We use RESISTANCE in RFI suppression and in common mode chokes.

The tutorial is at http://audiosystemsgroup.com/RFI-Ham.pdf

73,

Jim Brown K9YC



_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Jim Brown-10]

Hi Bob,

See comments interspersed.

On Mon, 29 Sep 2008 09:51:11 -0500, Solosko, Robert B \(Bob\) wrote:

>Jim,

> I've been reading through your tutorial, and, if I understand
>it, the best approach is neither a plain wound coax balun nor a string
>of ferrite beads strung on the coax but is a balun made with  coax wound
>around a ferrite core in a way the minimizes capacitance - is that
>correct?

It depends mostly on the frequency range and the ferrite material, and to a
lesser extent, on the power level. Think of the choke you're winding as any
other inductor that has VERY low Q (typically around 0.5), and thus a very
broad resonance. Like any inductor, we vary the number of turns, their
diameter and spacing, and the core material to hit the desired resonance. A
choke of #31 material can provide a strongly resistive impedance over a
frequency range of roughly 4:1, so we can wind one to cover 160-40 meters.
That choke would use closely spaced turns, because we need the additional
capacitance and mutual coupling between windings to move the resonance down
to about 80 meters. On the other hand, a choke to cover 20-10 meters needs
wide-spaced windings, because we only want to move the resonance to about 21
MHz. Power level enters the equation only to the extent that the choke must
provide sufficient common mode impedance that it reduces common mode current
to the extent that the P=I*E is small enough that it does not overheat the
coax or the core. The tutorial shows that's an easily achievable objective
once you realize that it's a key design parameter.  

> But I still have several more questions:

> - it appears that material 31 is the best material to use for
>baluns.

It's the best material to use for a COAXIAL CHOKE that needs to work below 5
MHz. #43 is equally good on 40M, and slightly better above 40M. If you're
only stocking up on one material and buying in quantity for the best price, #
31 is the best choice.

>In addition to my transmission line balun, I also have some
>problems with power supply birdies on 160m, and to a lesser degree on
>80m. Is material 31 still the optimum material for adding additional RFI
>filtering to my power supply (along with parallel capacitors)?

These chokes kill common mode current on the cable you wind around them, but
there can also be differential mode coupling that a capacitor ACROSS the line
(that is, plut to minus) can suppress, and there can be direct radiation from
insffficiently shielded circuitry. No external filtering will kill (or
change) that direct radiation.

> - My transmission line balun is serving two purposes: as a
>current balun to reduce the RFI in the shack problem that I have had,
>and as a 4:1 transformer to better match the ladder line from my antenna
>to the short length of coax to my rig. (My 4:1 balun is made from 2
>separate cores as is the BL2, but they're much larger cores to minimize
>saturation and heating problems.) It seems to me, if I understand your
>tutorial, that the characteristics of the cores used for transformers
>(low resistance) and those used for suppression (high resistance) and
>mutually exclusive. Thus, does it make sense to have a single balun
>serving these two purposes, or is it better to optimize the balun for
>the transformer application (material ??) and have a separate balun
>optimized for suppression?

That's a very perceptive question. For the first part of the answer, study
the photo of the high power DXE 4:1 balun in my Power Point presentation --
select it from

http://audiosystemsgroup.com/publish 

That "balun" is essentially two chokes wound with parallel wire transmission
line (that is, bifilar) around what could be #31 or #43 cores. On the 50 ohm
side, they're wired in parallel, on the 200 ohm side they're in series.
Because the chokes are bifilar, there's a lot of leakage flux in that core if
you're running power. The choking impedance isn't very high either. A far
better design would use coax, #31, and a lot more turns. That "balun" would
have a much higher choking impedance, and would also be a lot more efficient.

The second part of the answer is to study the DXE catalog -- they sell a
separate product that they call a common mode choke! When you study my
measurements (in the Power Point) for three of their high power "baluns,"
it's obvious why -- they have very poor common mode rejection.

So the short answer to your question is, YES!

Another point of clarification. Both the DXE two-choke series/parallel combo
and the one I described wound with coax are NOT transformers -- they are NOT
coupling signal through the core, they are using the core to form a choke.
They ARE doing impedance transformation and balancing. So it is correct to
call them "baluns" but incorrect to call them transformers.

73,

Jim K9YC


_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com