An interesting subject. I went into the QST archives after reading Jim's
post...
The first reference of any kind to "grounded grid" amplification appears on the
02/1948 issue of QST where the concept is used in a 6m tetrode amplifier. As
Jim pointed out, tetrodes were used extensively in GG applications and that
concept remained in a large number of amp projects that appear in QST all
through the 1950s.
In the 08/1961 issue of QST, Eimac's Orr, Sutherland, and Rinaudo discuss a GG
amp and mention the difficulty of measuring grid current. Their solution in
that project was to use a 1-ohm carbon comp resistor in the grid lead with an
RF bypass cap to the chassis. The voltage dropped across the resistor is then
used to report grid current.
But in the next QST issue (09/1961), Sutherland writes an article on what may
be the first implementation of B- isolation above chassis ground. See p. 14 of
the September issue for a PS schematic that resembles what we expect to see
today. Anyway, that's the first appearance I see in QST but B- chassis
isolation may have already been published in application notes by Eimac or
Machlett. Possibly B- isolation was published earlier in IEEE or even IRE,
its predecessor.
Regarding the joining of electrical circuit ground and neutral in an amp, that
practice lasted at least through the early 1980s. ETO's 70/77 series amps use
a 120VAC blower that tap one side of 240VAC and <gulp> the chassis. Through
the production life of the 77 which lasted into the early/mid 1980s, the effect
of such wiring is to ground the neutral in the equipment. That's not strictly
an NEC violation since the NEC is a premise wiring standard and not a products
wiring standard. Still, the practice would violate UL and other product
certification standards.
Thankfully, the fix is easy in the 70/77 series. One simply cuts a jumper on
the Cinch-Jones power plug. The 3-wire power cord is discarded, replaced by a
4-wire power cable. Not even the top cover of these amps requires removal for
the change.
Paul, W9AC
-----Original Message-----
From: Amps <amps-bounces@contesting.com> On Behalf Of Jim Garland
Sent: Saturday, December 12, 2020 10:57 AM
To: Amps group <amps@contesting.com>
Subject: [Amps] History of Grounded Grid Amplfiers
I received this interesting inquiry this week from Bob W4ZST about the history
of grid and plate current monitoring in grounded grid amplifiers. Bob writes,
/“Who came up with the idea of metering grid and plate current in the B-
circuit? I remember way back that many folks just put their plate current
meter in the HV line and (hopefully) protected the meter from access by a hand
or other body part. I remember some that had their meters behind a glass or
plexiglass panel. I was guessing that it might have come out of Collins but
haven't any info on that. If you know, I'd like to hear as it has just been
one of those things I'm curious about and haven't been able to find.” /
I don’t have a definitive answer to Bob’s question, so I’m polling the
group for comments and suggestions. Here is my personal historical take on the
issue: I built my first GG amplifier in high school in the late fifties (a pair
of 813s) and, like everybody else in those days, I grounded the negative HV
power supply connection, which was in a separate enclosure from the RF deck. I
remember that vividly because I almost killed myself when I unscrewed the coax
connector from the RF deck without realizing that the coax shield was the
ground return for the HV supply.
I dragged out my old 1956 ARRL Handbook for elucidation. Turns out the power
supply circuits there always show the negative power supply output tied to
chassis ground. Interestingly, the ’56 Handbook discussed the neutral line
appearing on “newer” house wiring, but the instructions were just to ground the
neutral wire to the chassis in an HV power supply. Many amplifiers of that era,
both homebrewed and commercial, did the same. Today, of course, that practice
would violate every electrical code in the nation.
In the 1950s, most HV power supplies used full-wave rectifiers with choke input
filters and the CT of the plate xfmr was grounded to the chassis. (I remember
reading about measuring plate current in the CT of the transformer lead, but I
don't believe that idea ever really caught
on.) Even then (say 1960 or so), nobody thought of lifting the B- return from
chassis ground. The Johnson Ranger transmitter measured plate current with a
panel meter in the HV line, common practice in its day, but a spectacularly bad
idea.
I think there were several technical advances which lead to the practice of
separating B- from the chassis. One was the growing popularity in the mid 60s
of grounded grid amplifiers. In those days, triodes like the 3-1000Z weren't
yet available, and hams usually just made do with surplus tetrodes, like the
813, 4-400A or 4-1000a. One could turn these tetrodes into makeshift triodes by
tying together their control and screen grids.
These new grounded grid amplifiers, presented several design
challenges: one was how to provide operating bias for the control grid when it
was securely anchored to the chassis. A workaround, (used in some Henry
amplifiers?) was to lift the grid to DC and bypass the RF to the chassis with
capacitors - a workable but clumsy solution that was also prone to instability.
Another challenge was how to drive a grounded grid amplifier with
tubes lacking independent cathodes. At first, low capacity filament
transformers provided the answer. These isolated the filament from the
transformer primary, allowing the filament to be driven with RF without being
shorted to ground through the filament transformer. That was another workable
but awkward solution that was quickly abandoned when ferrite filament chokes
were invented.
A big technical advance was the realization that cheap capacitor input filters
using series-connected electrolytic capacitors could replace expensive and
heavy filter chokes and their accompanying oil-filled capacitors. Once this
idea took hold, inexpensive voltage doublers in HV power supplies became
feasible. This innovation became quickly popular, since plate transformers were
cheaper to make without a center tap and with half the number of required
secondary turns. I remember struggling to understand how the voltage doubler
worked, since the circuit has no natural ground point. To me that was the first
time I realized that "ground" and "B-" were separate concepts, and that the B-
power supply return didn't have to be anchored to the chassis ground.
For me, that realization was a breakthrough, as it was for many others, and it
quickly led to today's practice of grounding a grid directly to the amplifier
chassis and lifting the B- above chassis ground by the operating bias voltage.
(The confusing issue to many builders still is why only grid current flows
between B- and ground, while only cathode current flows back into the power
supply. A related point of confusion is the polarity of the safety diode
between B- and the chassis, and how it protects the amplifier in the event of a
flashover in the high voltage circuit. Understanding current flow in a grounded
grid amplifier requires some careful study!).
I don’t know when this circuit innovation first appeared, but I
suspect it might have been the in the wonderful newsletters published by Eimac
and authored by Bill Orr (SK) W6SAI and Bob Sutherland (SK) W6UOV/W6PO. Maybe
somebody on our reflector has a more informed understanding.
73,
Jim W8ZR
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|