Amps
[Top] [All Lists]

Re: [Amps] Grid Dipping the Pi network in a new amp?

To: <amps@contesting.com>
Subject: Re: [Amps] Grid Dipping the Pi network in a new amp?
From: "Tom W8JI" <w8ji@w8ji.com>
Date: Tue, 17 Jul 2007 15:22:48 -0400
List-post: <mailto:amps@contesting.com>
> I'm more curious as to what your attempted power 
> parameters are.  If they
> aren't high enough plate load you will never be able to 
> find enough
> capacitance to resonate that thing.  I doubt you are 
> running 3  -8877's with
> 2500V on the plates.  They made them to run a 3-1000 zero 
> bias with 2500V
> and who runs just 2500V?  A 3-1000 isnt happy until it has 
> at least 3500V on
> it.  Even at 3000V that thing is already long for the junk 
> pile.

The nice thing about tank circuits is you can be off by a 
factor of  even three or more and it will barely make a 
difference in amplifier output.

All the worry about getting the Q right at 12 or some other 
magical number is largely a waste of time except the minimum 
Q results in less circulating current. All you really need 
to do is add enough C to make it hit.

Look for example what Dentron did in the MLA and 
"Clippertron". They had some fixation that a pair of 200 pF 
across a few hundred pF variable was about right. Switch 
them out on 80, and the C instead of being half the 160 
value is now nearly 1/4th the 160 value. The Q changes by a 
factor of nearly three or four, yet the efficiency remains 
almost the same. The amp doesn't suddenly "quit working" 
because the Q is not some magical number.


As a matter of fact many VHF and UHF tube amps work happily 
with a loaded Q of 50 or 100, sometimes even more.

Throw in a little more C and with the meaty coils in the B&W 
the thing will work just fine.

73 Tom


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

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