tubes age 12
paralleled a 35 mfd tantalum for the moderate fast changes, and then a 0.1
monolithic for the really fast changes, and so can't detect yoops in output when
changing the load drastically be scratching two wires together while watching
with a 200 MHz scope.
In the circuit I've sketched for a solid state 0A2, I include that capacitor from
output to input on the TL431. By the way if you use the DIP package the
regulator can be good for currents from about 0.1 ma to 100 ma, if you watch
the dissipation distribution between the 431 and the MRF820...
The plain 390 with 6082, 6AH6 and 5651 reference looks very amenable to
going solid state for the regulator. G3SEK has published a solid state regulator
for the screens of 4CX250 family tubes in QEX (also on his web page where he
offers kits and modules) that uses MOSFETs and should translate to the 390
easily with more heat sinking for greater power capability and probably a
current limit change. Removing the heater power of the 6082 is significant
progress in cooling the radio. Using a solid state regulator driver that puts 1
milliamp to the reference instead of 15 or 20 ma, saves considerable heating
also.
Changing the other stages is harder. The most applicable solid state
component is the double gate MOSFET. These are no longer made by US
makers, and are getting rare in Japan and Europe. Especially in packages
bigger than a grain of pepper. (I grind my pepper fresh so the chunks
sometimes are fairly large). Though yesterday I received a package from
Farnell Components in GB. They own Newark and MCM among others but the
local places in the US don't stock the same parts and Farnell has a supply of
BF-988 double gate silicon MOSFETs. They have a superb noise figure, lots of
gain, but are limited to 12 volts on the drain. I'll be applying them to my halfway
working solid state gear. The paper work weighed more than the parts, and it
cost 13 pounds (British money) shipping while 35 transistors cost only 10
pounds.
I'm kind of thinking that for replacing 6AK5 and 6BA6 with full supply voltages
still on the tube sockets, that a double gate MOSFET can act very much like a
pentode, with the gain controlled by the same negative grid bias, and with gate
two bias proportional to the screen voltage, and by putting the 0A2 voltage
regulator circuit in the drain, just need a tiny power MOSFET and the TL431...
And with that constant voltage drop bypassed with a small capacitor to make it
nearly invisible to RF. Might need to put a voltage divider on the signal gate or
an unbypassed source resistor to keep the gain within reason. The fanciest
silicon double gate MOSFETs can have a Gm nearly ten times that of the 6BA6.
And are good to more than a Ghz with the same gain so there might need to be
some ferrite beads on input and output to calm microwave oscillation
tendencies.
(14 pages)
(10 pages)
Manymanuals.com
Manymanuals.de
Manymanuals.fr
Manymanuals.it
Manymanuals.pl
Manymanuals.cz
Manymanuals.es
Manymanuals-pt.com
Commentaires sur ces manuels