[Amps] Heater chokes on 1/2" ferrite

[R.Measures]

On Sep 15, 2004, at 10:31 AM, Thomas Hoeppe wrote:

> Hello OM's,
>
> in most of the american handbooks they recommend a bifilar choke on a 
> 1/2" ferrite rod, when the tube's cathode is in direct contact with 
> one side of the filament or even directly heated. (3-500Z...) The 
> 8877, of course, has no need for such a choke.

/Thomas -- Unless the one side of a 8877/GS-35b/3cx800A7 heater is 
connected to the cathode, it is possible for a glitch to create an arc 
between the cathode and the heater.  Sometimes this arc will open the 
heater.  Thus, the safest course is to wire one side of the heater to 
the cathode and use a bifilar choke to RF-isolate the heater/cathode 
from the heater transformer.  In this case, both sides of the heater 
transformer secondary must float above circuit common.

> Did anyone check the inductance?

Roughly 10uH minimum is required for 1.8MHz operation. (XL=120-ohms)  
The +j120-ohms can be cancelled by adding c. 900pF to C2 of the tuned 
input for 160m.

> What is the permeablility of that ferrite material?
>
Typically 125

> I want to use a square rod with permeability of about 1000.

This should work well below 1MHz.

cheers

> It is a U/I shaped material. With 17 bifilar windings on the bottom of 
> the U, I got 24 uH, and when I close it by using the "I" I measure 230 
> uH. This is much more then needed on that point. Should I expect 
> problems? The input of the tube is around 50 Ohms (GS35b).
>
> I have already built a linear, which shows less gain on 160m then on 
> 80m. I am afraid some input power on 160 is lost in that choke. 
> Unfortunately I did not check the choke when I made that amp years 
> ago.
> Now, I want to avoid that problem in a new desktop project.
>
> Thanks for your comments....Tom, DJ5RE

Richard L. Measures, AG6K, 805.386.3734.  www.somis.org