Amps
[Top] [All Lists]

Re: [Amps] KK5DR's ferrite loaded

To: amps@contesting.com
Subject: Re: [Amps] KK5DR's ferrite loaded
From: Manfred Mornhinweg <manfred@ludens.cl>
Date: Fri, 22 Jan 2010 19:23:35 +0000
List-post: <amps@contesting.com">mailto:amps@contesting.com>
Angel,

 > You explained why a plate choke wound on a ferrite rod won't be
 > better than one wound on the usual ceramic rod.

I'm not saying that it won't ever be better. I'm just saying that a
ferrite loaded choke might not necessarily be the better solution in
every case! It's certainly possible to make plate chokes with and
without a ferrite core. Which is more convenient, can only be decided in
each particular case, by considering all implications, including cost
and size. A typical ham project might be best off with an air cored
choke, simply because it's the least expensive, while an application
requiring very wide bandwidth and small size, and that isn't very
cost-sensitive, might be best off with ferrite.

 > But on the other hand, a choke wound on a ferrite rod is used as the
 >  filament choke in the cathode circuit of almost every triode power
 > amp design. So it must work very well.

It certainly works well, with fewer possible pitfalls, in that low
impedance application. But the fact that almost every triode amp uses
one, does not mean that it is the only way to do things, and not even
necessarily the best way! Maybe it simply means that someone tried using
such a filament choke, found it to work well, published his design in
QST or the Handbook, and from then on everybody simply copied that,
instead of coming up with another and possibly better way of solving
that task! Ham radio technology is crammed full of such cases. Most hams
don't do electronic work. Of the few that do, most are copycats, their
expertise being enough to successfully copy a proven design, but not
enough to design their own and actually get it to work well. Only very
few do develop good new designs. And among engineers, there is also a
large percentage who prefer copying over doing their own original
designs! So it's no wonder that an idea that works well enough becomes
the standard followed by almost everybody, even if it might not always
be the optimal solution.

 > How is this? Is it because the cathode circuit is a low impedance
 > circuit and the plate circuit a high-impedance circuit?

Yes, this makes a large difference.

 > Also, the type of ferrite used for the bifilar filament chokes that I
 >  have seen in most published designs does not seem to be very
 > critical. Is it for the same reason?

There are two things here: First, the permeability in itself is totally
uncritical, because the inductance of the choke is dominated by the air
through wich the field closes. A rod with any permeability from 20 or so
up, will produce essentially the same inductance. With a closed core,
such as a toroid, this is different: In that case, the inductance is
roughly proportional to the permeability.
And the other thing is loss: Higher permeability usually goes along with
higher loss. So, it would be best to use a low permeability material.
But then, when you have just 50 to 100V of RF across that choke, and
have a enough turns, it will be working at low flux density, and thus
the loss might be low enough even when using quite lossy ferrite. In a
plate choke instead, the RF voltage is typically 20 to 40 times higher.
That would require 20 to 40 times more turns, to achieve the same loss.
And this might result in too many turns to be practical. But it has to
be evaluated in each particular case, considering the cores available.

 > Would a toroid be better than a rod for the filament choke?

Maybe, but only if it is of a low permeability material. A toroid would
give enough impedance with just a few turns on it. This is attractive,
because it results in low capacitance, short wire, small size, and so
on. The AC going through the filament is no problem, because it cancels
out in the bifiliar winding. But then you have the cathode current (at
least when the tube is directly heated, which it normally is). This
current would easily saturate a high permeability toroid!

A filament choke wound on a powdered iron toroid could be usable,
compact and inexpensive. On a low permeability ferrite toroid it might
be OK too. And on an almost closed ferrite core, with a suitable air
gap, it should also be fine. But probably not on an ungapped high
permeability toroid. Of course, the famous phrase applies: It has to be
evaluated in each particular situation...!

Jerry,

> I'm about to try one. I'll let you know if it melts.

Good! Let us know! ;-)

I don't think it will melt. But maybe it will have higher loss than 
another choke would. But I'm just guessing. Not having all details, I 
can't calculate with any precision how it will perform, much less try it.

Manfred.

========================
Visit my hobby homepage!
http://ludens.cl
========================

_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps

<Prev in Thread] Current Thread [Next in Thread>