Paul Christensen wrote:
>
>Regarding the Alpha PA-77...
>
>I just reviewed the schematic and here is my understanding of the
>transistorized grid over-current circuit:
>
>The cathode-side of a 8.2 V zener is coupled to the 8877's cathode through a
>15 uH choke and 0.02 uF bypass cap. The anode side of the zener is then
>brought to a discrete darlington transistor pair (Q4 & Q5). The darlington
>serves two purposes: 1) it functions as the electronic bias switch (EBS),
>and 2) the voltage dropped across the darlington's emitter resistor (R20 -
>3.3 ohm @ 1/2-watt) to ground is then sent to the base-bias resistor of the
>grid over-current transistor (Q2). The collector of Q2 drives a relay
>which in turn forces a manual reset of the amp into standby.
>
>If Q5 develops a CE short and subsequently opens R20, this could expose the
>base of Q2 through a 100-ohm resistor directly to the anode of the zener.
>Not sure what damage (if any) could develop from here. Anyway FWIW...
>--------------------------
>Parts used:
>
>Q2 - 2N5184
>Q4 - 2N5184
>Q5 - 2N5321
>
Thanks for providing some concrete information, Paul.
So the way it's normally meant to operate is that grid current flows
down through R3, and when the voltage drop exceeds about 0.6V it turns
on Q2 which trips the protection relay. That would happen at about 200mA
grid current, which makes pretty good sense (200mA x 3.3 ohms = about
0.6V).
Note that no grid current flows through Q2, and Q2 does not "interrupt"
grid current. It simply operates a relay which switches the amp to
standby.
What happens to Q2 in a major surge depends on the other things
connected to R20. If the circuit is at all conventional, R20 is the grid
current metering resistor, so there must be a meter connected across it,
and hopefully also a protection diode (anode to ground). Then the
B-minus rail and anode current meter connect to the top of R3.
R20 is in the path of any major current surge from B+ to ground, as it
flows up through R3 to return to B-minus. If the protection diode is
connected across R3, the voltage drop will not exceed say 0.7-0.8V, so
both R3 and Q2 should survive OK. Q2 *will* survive OK if the base
resistor is increased from 100 ohms to say 10k (and maybe change Q2 for
a device with higher current gain... I don't know the 2N5184).
So, as usual, the survivability of the DC/control circuit - and the
meters - depends on getting several design features right.
NONE of this is any excuse for disabling a grid current protection
circuit! Yesterday I tuned up a 3CX800 amp I've been restoring. Since I
don't have four hands, three eyes and the reactions of a fighter pilot,
the grid current protection operated about five times before finding the
optimum combination of the tune, load and drive controls. No problem - I
knew that whatever I did, the grid would be in no danger.
As for grid fuses (either control grid or screen grid), think what
happens when they blow? If they let the grid float free in the electron
stream, that puts the whole tube into even more danger.
--
73 from Ian G3SEK Editor, 'The VHF/UHF DX Book'
'In Practice' columnist for RadCom (RSGB)
http://www.ifwtech.com/g3sek
--
FAQ on WWW: http://www.contesting.com/FAQ/amps
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
|