Cathode biasing a pair of EL34s

[dude]

I put a bias switch in one of my 6v6 Plexi’s, fixed to cathode. I used one 250 ohm cathode R. My dissipation at idle is 18.4 watts. Impedance of OT is 3500ohms, using 7k OT @8 ohms but set at 16 ohms (3k5). I’m not sure of the rating of the Hammond AO43 PT but l’m using a solid state diode plug-in the tube socket, less current draw on PT. Nothing seems to get even warm. Plate voltage warmed up at idle on EL34 is 380, screens about 15 volts lower. The amp sounds much better on the cathode setting, lots of compression and some nice wanted sag, l really like the tone. l have no PPIMV, just pre-amp master, 220k grid leaks.The fixed bias setting is a totally different sounding amp, the EL 34’s are set to 70% bias. Amp is more Fenderish sounding, more headroom, way less compression but useful.
My question is even though l’m running cathode biased, the amp is still in “AB class?” So, the 250 ohm cathode R @ 18.4watts dissipation is about 70% of the max output for an EL34, 25 watts. Am l good to go? From my experience cathode biased 18 watt amps with EL84’s run at 9 watts idle dissipation or 100% tubes max, l assume these 18 watt Marshall’s are running class A, ? but the Plexi is AB class, (cathode biased) so only 70% dissipation. Forgive my ignorance but do l have these assumption correct?

[shooter]

is still in “AB class?”

normally self biased only gets to A1 since it's always conducting during the full sinewave
I typically run Class A at 90 to 105% depending on sound, tube life isn't considered in self biased  guitar amps 

[dude]

So, does that mean this Plexi switched to cathode biased is now running class A1 and should be close to  90%/100% of tubes max dissipation? Does this mean dissipation at 18 watts is cold biased, and these EL34’s should be self biased closer to 25 watts? Lower the resistor to say 200 ohms from 250 for two EL34’s...?

[shaun]

You may be confusing the 18watt designation. They're talking about rms watts, I believe, not tube dissipation watts.

With a "fixed bias" amp - which is actually not fixed in the same sense a cathode biased amp is - the grid current is controlling bias of the tubes, so because that current fluctuates considerably, you need more head room. Thus, the 70% recommendation.

It's commonly held that cathode biased amps can be run up to 100% of their rated capacity. 6V6s and - I think - EL84s are often rated at 12watts max - it will be listed in the manufacturer's spec sheet. Some believe it's better to run them at anywhere between 70-90%. So get your iphone and figure that. Mine tells me that 70% of 12 is 8.4 watts. For a hotter bias - which may effect tone and your ear - use a lower value cathode resistor.

As far as Class A versus AB, I dunno. But check the data sheet for an EL34. Look for the max diss rating, and choose a cathode resistor(s) that gets you close to what you want. I just looked at some EL34 datasheets that said 25-30w max. 70% of 25 is 17.5 - maybe that's the 18watts you meant? Anyway, I'd aim for somewhere around 20w per tube if it were me - tubes will stay cooler and last longer. And hotter doesn't always mean a better sound.

Forgive me if you know all this, just trying to help.

[shooter]

The amp sounds much better on the cathode setting

the statement you made, is my 1st criteria once i'm "close"
I would play for a week, then decide if it's worth tweaking because you will find a demised return, as tube current goes up, plate volts come down, current down plate V up and the only real evaluation is sound (assuming math is reasonably close)

[sluckey]

Lower the cathode resistor to 130Ω to 150Ω. This Matchless uses two 270Ω resistors...

     https://el34world.com/charts/Schematics/files/Matchless/Matchless_clubman.pdf

[dude]

Yes, Sluckey l think l’m getting it, running cathode bias changes the class of operation from AB TO A1 as Shooter mentioned. In my case running my 6v6 Plexi, cathode biased is classA1, so dissipation can be much higher than in fixed. With EL34s l’m running them at 18 watts dissipation not sure if l took the screens in consideration..?
What started me on this quest for tone was l lost a few sales to the new Marshall 20 watt Plexi head copy, a cathode biased EL34, 20 watt head, pic’s shows the Marshall head has a single 150 ohm R, so you’re on target with your suggestion. I assume they’re pushing that Marshall head pretty hot, the reviews are very good. I’ll throw in a 150 ohm, should raise the dissipation to around 22/23 watts. Does this mean l now have a 45 watt amp? The plate voltage with the lower cathode R, will drop, l assume cathode biased  l’m probably around 22/25 watt amp?

[tubeswell]

My question is even though l’m running cathode biased, the amp is still in “AB class?”

Class B is determined by the ability of the voltage swing at the output tube grid to achieve a negative-enough voltage swing that puts the output tube into cutoff. There's 2 ways of getting there.
1) Either the EL34 bias voltage is low* enough viz-a-viz the driving voltage (from each PI plate), then a big swing from the PI plates will put alternately put each side of the output stage into cutoff.
2) Keeping a bigger bias voltage, but having an even bigger PI output swing can also achieve this. 

*i.e. = 'hotter bias'


This can be achieved with either fixed bias or cathode bias output stages

[dude]

Parallel a pair of R’s with different wattage ratings, which rating applies. 250 10 watts with 350 5 watts, what the the watts rating?
Thanks Tubeswell, good info

[shooter]

Does this mean l now have a 45 watt amp?

you can answer that after re-bias, happy with tone, then;
inject a sinewave, measure VACrms at speaker, keep bumping up volume til signal quits getting bigger, square the reading and divide by speaker impendence = audio power out

[dude]

No scope,