I'm a push-pull animal, and this is my fave phase splitter (I take output off both plates) since it has gain, and its balance may be adjusted by the values of the two plate resistors.  It occurs to me that if output is to be taken from the grounded grid side only, perhaps the load (plate) resistor on the driven (by signal) side, or the actual common cathode (this is another name for the plate-loaded amp) side may be omitted.

This is another use for this circuit that I had not perceived. I have used it as a preamp as shown here:

This uses transformer output to reduce the gain and Zout, rather than negative feedback, which I have found to be a technically attractive but sonically dangerous tactic.  The problem with this circuit is the output transformer, which is unobtanium.

Okay, now I'm seeing what is happening here. Thank you for your very clear explanation of this circuit.

Yer Obeediyunt Servunt,

Poinz

The 6B4G SE amp currently in my system is 6n1p (both sections parallel) driving a 124B IT (parafeed). Talk about el cheapo, the output transformer on that beast is a Rat Shack 70v line transformer!

TC

Z out and gain is moderate (depending on the tube) but the circuit is really easy to drive and really wide-band. People rave about the Rosenblit version. (All 12AU7s, I think).

I like parafeed output circuits, they seem to swamp the loudspeaker resonance so they are much more tolerant of load. I guess there is probably some trick about setting the Q of the parafeed cap and transformer to keep it from getting tubby, but that's probably in the sizing of the components. I don't know, I just know I like what I've heard.

I guess the Radio Shack 70V transformer work pretty well, and the price is right too. That's cool. Since you're using a common distribution transformer, can you get substitute parts from other companies too?

Actually, to AC ground the 'on' side plate, all that's necessary is to remove the load resistance, since the next thing up that wire is a big old capacitor to ground - significantly larger than 10µF, usually.

Thanks again,

Poinz

Little peanut cheapos that sound like sh, ah, not so good.

Great bloody whacking things the size and expense of output transformers.

TC, my first try in the circuit above was the Hammond 124B.  I really wanted to like it, but it sounds just like the 10K outputs I tried in the 6V6 amp; dry, opaque.  I even tried Amveco power toroids (115+115 : 44), which Bench liked, but they were rolled off too much on both ends, and sort of wiry sounding. The Lundahls and Sowters and Tangos are all these ten watt monsters that cost $100-200 and won't even fit in the chassis.  It seems impossible to get the quality without the huge capacity.  Don't know why they feel that a component that's transforming a few milliwatts has to be rated at ten watts; maybe a traditional use for them is driving big transmitter tubes like 845's in A2.

I'm still looking, and also talking to some OEM winders who may be willing to wind to spec, but so far no joy.  I'm not giving up because I have a long affectionate history with the singly-driven diff-amp (thus my original Holy Toledo!). The high PSRR of diff amps makes them a natural for low level signal use; don't have to use huge music-sucking caps in the power supply final stage.

Poinz

There's a small chapter in the Radiotron Manual (3rd ed.) on parafeed and sizing the cap to avoid resonance with the L of the transformer primary. Basically, it's easy, 'cause bigger is better, say 1-5 uf, some use up to 10uf. I use 400-630v Solens or vintage paper in oils. The Radiotron manual also mentions that one may want to intentionally size the cap to resonate to "increase bass response"...

There's also the Rp of the tube to consider and the quality of available caps!

The Rat Shack line transformer works great as a parafeed output - I'm sure something from allied or edcor would work just as well. Plan on experimenting with the various taps. I want to try using a PP OPT in parafeed.

BTW, I'm using the primary of the original Hammond 1627 opt as a plate choke.

Anyone who reads my posts, remember, I'm no EE or anything - I barely have the math to do anything correctly. Just an experimenter.

TC

I just started making breadboards last year.  It's a revelation; I don't have to worry about drilling extra holes or beating them up.  I've gotten more developmental prototyping in in the last year than the previous five.

Poinz