One of the most dramatic improvements to me was to fix the power supply so that the screen of the 6L6 was much closer to the plate voltage, the factory setup with low screen voltage really was wimpy, but maybe it was intentional to keep the temperature down.  It does have a good OPT for Class A single 6L6GC use with an 8ohm load, too bad they put it inside the chassis (near the reverb *tank*), it gets too hot when you run the 6L6 at what I consider more reasonable Class A dissipation now.  So I always run a small desk fan behind the amp, that way it runs cooler than a regular Champ.  Well, one of the helpful things also was to replace the old 6L6GC with a NOS JAN 7581A which can take the heat just fine and really rocks so naturally I pushed it to the max :)

500V B+ and 250V screen were intentional, what were they thinking?  If they were thinking, it was probably about compromising the things I value most in a Champ.

When I worked on the power supply and got the screen closer to the plate voltage, it made my guitar sound a little more recognizable.  I did not lower the plate, I increased the screen, and the plate declined slightly as a result.

This also brought the 6L6 operating point more into the area where Class A operation was possible without sacrificing most of the power the 6L6 is capable of.

Naturally with the dismal nature of this design, the other hardware is not fully capable of supporting a tube as big as a 6L6 if you are trying to get nominal performance from an NOS tube.  I upgraded the resistors that were unsoldering themselves, plus tweaked the power section and fixed the *designed to hum* grounding problems, and basically felt I had done all I could with the 6L6 and the amp was still a dog.

Back in the cabinet, the SE6L6 does run too hot for the PT, and I'm OK with the OPT inside the chassis WRT noise (if its done right, Champ12 is not TOO bad), but it contains the heat too much, so that's the big problem with trying to get nominal performance from a SE6L6.  I made it so this amp can BARELY handle playing in 90 Fahrenheit room temperatures without a fan, but I still always use a desk fan on high blowing at the 6L6 and the chassis in the area of the PT even when the room is below 70degrees.  This is better (cheaper, quieter and more easily replaced at WalMart) than a muffin fan would be, it also ventilates the entire open back cab.

A)      The original tube design data will not apply to this amp since the important options are no longer available unless you are willing to change PT to a different voltage.  I just accepted what the PT was and tried to make the most of it.  Real champs have the screen about equal to the plate at idle and this is what I like.  I like 500V B+ for NOS 6L6GC, even though you can't really get to the target operating point for Class A, you're still SE and there are advantages that make it a workable balance.  Must have experimented with dozens of SE6L6 circuits on the bench at about 500V before I worked on the Champ12 to begin with.  It's cathode bias anyway so most charts for fixed bias are not directly applicable.  Bigger increase in output as the screen goes above 350 toward 500V.

B)      The 250V screen limits dissipation, power, & tone of the tube overall, you could add a 1K screen stopper that will effectively limit screen dissipation to prevent gross tube damage under most conditions, this is not necessary with only 250V and I didn't do it either with the arrangement I used after changing the screen voltage.

C)      The plate voltage is easily tolerable by NOS quality tubes, especially with the screen only at 250V, once you get the screen up to a decent point for guitar use, the the thing to worry about is the plate dissipation which will increase dramatically, so you have to make sure the cathode bias resistor compensates and the tube is not hot enough for the plate to reach a red-hot temperature.

D)      The tube's voltage is supposed to be measured from plate to cathode, this means fixed bias amps are subjected to more than cathode bias from the same power supply.  You would use the voltage drop from plate to cathode to figure the heat dissipation of the plate itself based on how much current is being conducted by the plate circuit.  Sometimes I imagine *maximum plate voltage* might be specified absolutely to avoid arcing between octal pins of diverse potential, but the type of tubes I like exceed the specifications well enough for this to be a non-issue for 6L6GC at 500V.

The power supply was much easier than the main PCB, here is my hacked result:


Here's some info from my notes, its difficult to be sure of everything
without opening the amp, but I think I've got it covered, you'll have
to look at your layout carefully if you want to end up with a more
roadworthy wiring plan, some of my connections are in mid-air, remember
to always remove the tube before soldering on its socket:

new 22/350 cap
C+ at pin 4 is removed and soldered to negative of 22/350, its
 positive is to pin 4.

new 6K 3watt resistor

about 1.5 inches of red wire is added to the B+ terminal on PCB,
 then solder the other end of the red wire to the 6K, and the remaining end of the 6K solders to pin 4 along with the above cap.

Change preamp dropping resistor:

R51 is removed and replaced with 1.5K

New cathode resistors:

R32 & R33 are removed and replaced with 400ohm 7watt cement type,
 leaving long enough leads so there is much space between the hot
 resistor elements and the PCB.  These resistors were soldered so
 they were on the opposite side of the PCB than the original parts
 were.

R31 is removed and effectively replaced with 250ohms 10watt cement.
 The new R31 is placed against the chassis (for heatsinking) near the
 OPT then wired to the PCB with stiff solid core insulated wire.
 Then pin 8 is wired to the other hole where the old R31 was on the PCB.

(I believe I only soldered one end of the new R31 to the PCB and the other end directly to pin 8 (maybe with a little extra wire), then had to run a wire from pin 8 back to the PCB to make it equivalent to simply replacing the resistor, I had probably removed the connection to pin 8 before I documented how the replacement was installed.  There is also a note that the only chassis ground I left was at the green
lug from the line cord, and from there I left only a single ground wire
 going to each PCB.

Also it says *much solder is added at junction of black grounds on PCB*,I think this was a conductivity improver and lowered the hum a little.

dave, The Hammond 270DX is 275-0-275 rated at 90mA with
3mA for the tube rectifier. You could use a 5AR4 and get around
355v or a 5U4 for probably close to 340v. The 6L6 only draws
66mA in class A according to the RCA Manual. I think what's
throwing everyone off is the "...in a champ 12?" question. Are you
going to mod the champ? Drill? What do you have in mind exactly?
 
I went 'round and 'round with this for the past year. The biggest
problem in my mind is that there is no real selection of SE OPTs for
guitar amps. I wanted to try an EL34 or a 6L6 amp myself, but I
wasn't about to spend around $100 for the Hammond 1627SE for
7 watts I might not even like. I finally ordered one. I've got three JJ
KT88's coming with it. A nice bonus I've just discovered is that I
can use a 6L6. The B+ is right for it and I can adjust the fixed bias
for it. RCA Manual calls for 4200 ohms ideal load on the primary,
so I'll just use the appropriate tap on my 2500ohm tranny to get
5000ohms. Close enough for rock 'n roll.
 
John