Another 6L6 schematic

[sorenj07]

My only concern is that the 6N1P (amplification factor of ~ 33) might not be able to swing enough voltage for the 6L6 tubes which I bias at around -35V. Anyone have any opinions? I have all the parts to make this amp otherwise, just haven't bought the actual resistors and capacitors.

Note: I forgot to show the ground between the two output stages, and the output transformer is Edcor's 6600/8 ohm 60 watt model.

[sorenj07]

I guess I should rephrase the question: will it work? I'd prefer not to drop around $40 on parts including delivery and more on a chassis and have to buy new parts and/or "undrill" the chassis...

[SY]

I haven't laid out a load line to see if you can swing the 70V peak to peak, but for sure you don't have even close to enough gain. And the short tail guarantees imbalance. A 6L6 will almost always require two drive stages or more.

[sorenj07]

Thanks, at least I know now. Is there any way of calculating voltage swing and gain from tube curves? Time to re-think this one. I do have two pairs of 6SN7's I might try, a la Williamson

[SY]

The swing and gain can be best gotten from load lines on the characteristic curves.

[jon_010101]

Hmmmm, good points from SY. I think the gain may be passable with ~3-6dB feedback. mu~37 for the 6N1P if you bump up the current to 8-10mA, which would be very worthwhile. 120V on the tail would be nice -- the plate loads could be reduced slightly, so you can pick up some voltage there. 120 across tail and plate loads, can we get an extra 20V on the supply for 165 across the 6N1P? (looking at curves) At 8mA, that would give a tail ~7.5k and plate load ~15k, which is still a bit over 2x the plate resistance, and a bias of -2V .

If you want to be able to apply a good amount of feedback (~20dB), you could just tack on another stage direct-coupled to the phase inverter (like a Mullard amp) -- makes for easy biasing of the LTP. Maybe use a 6DJ8, 6N1P, or 6SL7 gain stage into a 6SN7 driver LTP. Or a 6DJ8 into a 6N1P driver LTP, etc. etc. etc. Many combinations

[SY]

Problem is that as a diff amp, you lose half the gain right there. So there's no room for feedback. Nope, there's gotta be two stages if triodes are the driver/phase-splitter tubes.

[EC8010]

Where's the differential amplifier? I see two common cathode amplifiers, one driven by the input, the other driven by the output. Those cathodes need to be tied together to even approximate to a phase splitter.

[SY]

The two cathodes are tied together at the bottom of their 1k bias resistors through the 11k tail. How about "diff-ish" amp?

That brings up another fatal flaw, the way the feedback is applied.

[EC8010]

Whoops, missed the 11k resistor. But the feedback connection holds that grid at 0V... "If I was going there, I wouldn't start from here."

Tweaking a Williamson (or adopting it wholesale) is more likely to bring success.