[Amps] 4CX1500 class AB1 vs. AB2

[Gary Schafer]

> -----Original Message-----
> From: Tom W8JI [mailto:w8ji at w8ji.com]
> Sent: Thursday, November 02, 2006 11:17 AM
> To: Gary Schafer; 'Ian White GM3SEK'; amps at contesting.com
> Subject: Re: [Amps] 4CX1500 class AB1 vs. AB2
> 
> > If there is severe overshoot coming out of the exciter any
> > amplifier is
> > going to have problems with it causing distortion
> > products.
> 
> A grid driven tetrode has the most severe problems with IM
> perfromanc. Although they could be designed with controlled
> negative feedback, most are not. That being the case many or
> most are already poor as a general rule compared to cathode
> driven PA's. Case in point, look at the grid driven
> commercial amateur amps on the market recently.
> 
> Grid driven amps are especially intolerant of occasional
> overshoots when they have loose goose bias and screen
> supplies that have slow recovery times.
> 
> > With an AB1 amplifier with a stiff bias supply how do you
> > keep from
> > destroying the grid unless you NEVER bump into grid
> > current? With a higher
> > impedance bias circuit it is self limiting in grid
> > protection and alc is
> > easily developed as is done in the Collins 32S1 etc.
> 
> I think you mean the 30S1. The 30S1 is OK because the very
> high amounts of negative feedback caused by cathode drive
> overcome the poor design of loose screen and grid bias. The
> NFB stabilizes the exciter load impedance and corrects some
> of the distortion, so overall it is pretty clean for a
> poorly regulated design. It was also a very good match for
> the exciter power, and many of the rigs back then didn't
> seem to overshoot as bad as they do today. It also had
> pretty fast voltage recovery on the grid after being
> overdriven.
> 
> Take away that negative feedback and the case changes.
> Especially when the leading edge overshoot is handled by
> allowing the grid bias to bump way up.
> 
> I'd be careful about using grid impedance to protect the PA.
> As a matter of fact, I'd never consider it a good idea.
> 
> 73 Tom
> 

No I did mean the 32S1, KWM2 etc. but the 30S1 falls into the same category
as alc and grid protection are derived in a similar manner in all. A
moderate grid impedance protects the tubes from excessive grid current.

Negative feedback has little to do with being better at reducing the
distortion produced by running into grid current than having no feedback.
Sure an amp with feedback is going to lower the overall distortion X number
of Db but on a relative scale overshoot problems are going to create the
same amount of difference in distortion products whether there is feedback
or not.

The question still remains, how do you protect the grid of a 4CX type tube
from excess grid current with a stiff bias supply?

73
Gary  K4FMX