[Amps] RE: 4CX250B Screen Supplies

[R.Measures]

On Jul 24, 2004, at 7:42 AM, Steve Thompson wrote:

> On Saturday 24 July 2004 12:05, R. Measures wrote:
>> On Jul 23, 2004, at 7:47 AM, Steve Thompson wrote:
>>>> Also, with a 2-tone test, it would seem not to matter whether the
>>>> screen-V is regulated or unregulated since screen-I is constant and
>>>> therefore screen-V would not vary unless the electric-mains V
>>>> fluctuated.
>>>> end
>>>
>>> Doesn't the screen current vary as the rf output varies from zero to
>>> peak at
>>> the beat frequency between the two tones?
>>
>>   Hello, Steve ---  Yes, because there is zero screen-I until the
>> instantaneous anode-V dips down close to the screen-V -- i. e., the
>> anode is no longer positive enough to take away all the electrons that
>> pass by the screen.   For instance, with an 8171 @ E-screen = 1500v,
>> during the RF cycle, the screen current varies from zero to about 
>> 900mA
>> even though the average screen-I is c. 120mA.  From the characteristic
>> curves, it can be seen that most of the screen-I flows when the
>> instantaneous anode-V dips under 2kV.
>>
>>> The meter reading will be steady
>>> because of inertia.
>>
>> The screen's RF bypass-C keeps the instantaneous screen-V constant
>> during the RF-cycle.
> I'm not talking about the screen voltage and the RF cycle - I'm 
> talking about
> the screen current and audio frequencies that arise from the two tone 
> test.

RR.  I Understand this,
>
> If I wind the carrier power up and down, the screen current goes up 
> and down.
> If I wind the power up and down at 2Hz the screen current goes up and 
> down at
> 2Hz and the meter needle will follow it. If I wind the carrier power 
> up and
> down at 1kHz the screen current goes up and down at 1kHz but the meter 
> needle
> indicates steady 'average' current because it can't follow the 
> changes, not
> because the current is actually constant. The regulation of the 
> voltage at
> the audio frequency does matter. It's tricky winding the power control 
> up and
> down at audio frequencies manually, two tones does it more effectively.

   I agree, Steve.  However, unregulated screen supplies have a filter C 
that is large enough to minimize 100Hz - 120Hz ripple, so that should  
minimize V-out ripple (and IMD) when the load current changes at a 
1000Hz rate.
   In the amplifier I am currently building, I will be using an 
electronically-UNregulated resonant-choke 1500V screen PS.  The 
resonant choke is 9H//0.2uF and a filter C that is seemingly too big.  
The logic here is that since both of my previous "plywood box" ugly 
amplifiers used electronically-regulated screen supplies -- which 
proved to be the most delicate part of the machine, something simpler 
might be worth a try.  Thus, extra filter-C is an attempt to minimize 
voice modulation IMD.  With luck, the result won't suck.  If it does, I 
will go to a higher tap on the transformer and electronically regulate 
down to 1500V with the + floating regulator shown at
http://www.somis.org/D-a-08.GIF cheerz
end

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