wrote:
I was browsing through Terman (2nd edition) last night and read a
paragraph about "Dual Grid and Class B triodes". According to Terman,
if two concentric (but similar pitch) grids are put in the triode and
then connected to each other as a single control grid, the result is a
high-mu triode that needs zero bias for Class B operation.

I do not see dual-grid tubes mentioned in Terman's 3rd edition in the
same way.

Googling the term, I see that the 52 tube seems to be an example where
both grids are brought out to individual terminals.

I also see that sometimes "dual grid" is used to describe RF tubes
where there are two pins for a single grid (to decrease inductance I
guess), I'm not talking about these tubes.

The two-grids-connected-together characteristics remind me a lot of,
for example, the 811A (the most familiar Class B triode I'm familiar
with), but that only has a single grid terminal. Am I correct that the
internal grid structures of an 811A are essentially that of two
connected grids? If not, what inside an 811A makes it zero-bias high-
mu class B triode, as opposed to say its externally similar non-
identical-twin the 812A (a low-to-medium-mu triode that needs bias)?

I also note that Terman claims that the dual-grid structure forms a
very good electrostatic shield between heater and plate, and see that
811A's are often used in grounded-grid connection in RF amps. (Must be
a bitch to neutralize in common-cathode).

I've been intermittently playing around with SPICE to model 811A
curves (including grid current at positive grid voltage) and none of
the conventional triode models work right at all - its curves are more
like a pentode (in fact it's pretty trivial to fit it this way if you
let the diode characteristics take over at low plate voltage).

Tim.

Dual-grid tubes were popular for a short while. They were touted as
being versatile -- connect grid 2 to grid 1 and you had a high-mu
triode, connect grid 2 to the plate and you had a low-mu triode.
Judging from my tube data books, it didn't take folks long to decide
they just wanted one or the other, not two in one package.

The curves look pretty triode-like to me, once you take into account the
fact that the thing is running almost exclusively at positive grid
voltages. Perhaps the SPICE triode models aren't taking this into account?

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Posting from Google? See http://cfaj.freeshell.org/google/

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" gives you just what it says.
See details at http://www.wescottdesign.com/actfes/actfes.html