Amps
[Top] [All Lists]

[AMPS] L7 Amp Questions

To: <amps@contesting.com>
Subject: [AMPS] L7 Amp Questions
From: W8JI@contesting.com (Tom Rauch)
Date: Sat, 23 Sep 2000 19:00:09 -0400
> Another question is on Drake's method of grounding the grids using RF
> chokes and shunt capacitors to ground. The ampss that I've seen in the
> Radio Handbook ground the grids directly. What's the
> advantages/disadvantages of doing it the Drake way? Any harm in just
> grounding them directly? 

I believe that idea came from Bill Orr, as an offshoot from the 30S1 
Collins PA. In the 30S1, the control grid (the tube was a tetrode) 
was grounded through a high impedance for RF. The control grid 
floated through a capacitive divider with a high shunt impedance.

When the grid would be driven into grid current, the low cathode to 
grid impedance would pull the grid up to cathode potential, 
preventing damaging grid current. During normal operation (with no 
grid current) the grid operated at a voltage division between the 
cathode and chassis while the screen saw full excitation voltage.

Unfortunately, this system was misapplied to triodes running with 
grid current. If you look at what happens in a 3-500Z, the grid 
impedance varies all over the place in normal operation. This 
causes uneven feedback, and reduces gain flatness. 

The 3-500Z also has no shielding screen grid, so the addition of the 
grid to chassis reactance increases unwanted regenerative 
feedback at low frequencies. The added series inductance at very 
high frequencies results in extra unwanted feedback at upper VHF, 
making the tube less stable there also.

It's a poor system, but "monkey see monkey do". Once one 
manufacturer did it, the others followed. Bad idea or not. 
 
> And yet another question. I believe that one of the tubes is OK so I
> really only need one tube for the amp. The write ups I've seen on using
> two of these tubes says they must be matched. How closely do they need to
> be matched? Does one just check for idling current and verifies that each
> tube draws the same current +/- some percentage, does the grid current
> have to match within some amount, or what?

Tubes only need to be reasonably close. I wouldn't mix brands, but 
there is probably little reason top go out of your way to "match" 
tubes. The only real problem would be if you have a weak tube 
mixed with a "good tube".


   
73, Tom W8JI
w8ji@contesting.com

--
FAQ on WWW:               http://www.contesting.com/FAQ/amps
Submissions:              amps@contesting.com
Administrative requests:  amps-REQUEST@contesting.com
Problems:                 owner-amps@contesting.com


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