[ARC5] AM linear amplification

[Robert Nickels]

On 12/19/2016 9:27 PM, AKLDGUY . wrote:
>
> Class B is interesting. It's not widely known that the 807 / 1625 is 
> capable of Class B zero bias operation, with screen grid connected to 
> control grid*. This configuration eliminates both the negative grid 
> supply and the screen supply.

Tying the grids together treats the tube like a triode.   There's 
another class B configuration in which the control grids are grounded 
and the drive is applied to the screen grids instead. This circuit also 
eliminates the bias and screen supply components, and was used by Gonset 
in the modulator for the G-77 mobile transmitter, with a pair of 6DQ6 
sweep tubes as the modulator.   I got curious about it and did some 
research and found hat the circuit was patented in 1944 and assigned to 
Guided Radio Corp.   US patent number 2,358,148,  which was allowed 4 
claims.

One claim was based on the higher input impedance when driving the 
screen rather than control grids:

"The increased impedance results in greater independence between the 
plate current and the plate voltage when excitation is applied to the 
control grid  and it is possible for the plate impedance to vary 
considerably without introducing serious distortion or loss of power. 
Also, a much higher plate voltage can be applied to these tubes without 
exceeding the limits of normal plate dissipation than is possible when 
the tubes are operated as beam power tubes."

Another claim was based on longer tube life due to reduced heating and 
improved efficiency.

The inventors also claimed the distortion levels in this circuit (using 
a beam power tetrode) are "substantially less than obtained with zero 
bias class B operation or with beam operation" and say the circuit is 
"substantially linear even at low input levels"... "similar to those 
obtained in the operation of class A triodes". I'd read that distortion 
at low signal levels was a limitation of the triode-connected tetrode 
amplifier.

Around the same time I had a Multi-Elmac AF-68 on the bench and since I 
was replacing driver transformer anyway, I first wired it up as per the 
Guided Radio patent, with audio drive to the screen grids,  no bias,  
and the control grids grounded.   It worked and sounded fine with 
equally good fidelity, but I wasn't able to get above about 85% 
modulation with the existing driver stage.   I put it back to the stock 
configuration but this proved that the grounded-grid tetrode class B 
amplifier is well worth exploring. The G-77 schematic is available on 
BAMA for anyone who'd care to have a look.   Others may have also used 
this circuit but it was a new one to me.

By the way, Guided Radio was in the business of providing what were 
called "divers radios" which weren't radios at all but rather what we'd 
call an intercom for two-way voice communication between a tender and 
submerged divers.   So it made sense that the company would be 
interested in efficient audio power amplifiers.

73, Bob W9RAN