Cathodyne Mu Follower Line Out

[ruffrecords]

Overall the circuit does not make sense to me. What is the objective? A buffer, no gain or a buffer with some gain?

It has a LOT of power gain. That is the point.

Cheers

Ian

 

[Cqwet Dbdfte]

Ok, just a buffer.

 

[ruffrecords]

Ok, just a buffer.

Not just a buffer but a buffer with a high drive capability balanced floating output. And a cost effective one. It does with one triode what most buffers with a similar power gain would take two triodes to achieve.

Cheers

Ian

 

[merlin]

It does with one triode what most buffers with a similar power gain would take two triodes to achieve.

By replacing one of those triodes with a MOSFET

 

[Cqwet Dbdfte]

Maybe like this...?

R18,19, 20 may likely have to be adjusted to cover FETs parameter spreads.
Much smaller than a tube circuit.
Besides the output cap and R20 it could be a postage stamp size deal, with not much heat generation. Circuit is adapted to 48V limiting output swing before clip to 20Vpp.
J1 could alternatively be another DN2540, with a few resistance changes.
The load could of course be a 1:1 transformer.

 

[atavacron]

By replacing one of those triodes with a MOSFET

We could get really economical and just use a single power mosfet instead of the triode. Again requiring a well filtered supply. STF2N80K5 looks pretty good as a drop-in replacement, with fixed bias — 1/4 of B+ plus 4V. All diodes included, shrouded heatsink, low input capacitance (~100pF) right in the useful Vds range for us. Dunno how to extrapolate gain from the data sheet, but would be a pretty-frickin-low Zout regardless. 3 ohm Rds(on), suitably low, but might not matter here. Pretty much the same standards as choosing a source follower.

Maybe with a trimmer in series with the drain to get the signal balance right. Dicier than trimming the source resistor but at least you’re not changing the current that way. Strangely, some “source-o-dyne” schemas I’ve seen have a *lower* ohmic value from B+ to drain than source to ground, the opposite of Merlin’s example above.

 

[atavacron]

Maybe like this...?
View attachment 129116

I like how you’re thinking about this, but the idea behind the phase splitter and 2:1 trafo is that driving 2400 ohms requires way less current than driving 600. And it needs to work with voltages you’d normally find in a tube amp. In my post just prior to this one, I’m thinking B+ of 160V, because then the decoupling cap at the primary could be a 100V MKS-4, which is not unreasonably large. But other than that one part I’m with you on the postage stamp aspect for sure.

 

[Cqwet Dbdfte]

The DN2540 circuit can be adapted for that too, as it handles up to 400V. A heatsink would be necessary.

 

[Cqwet Dbdfte]

...or like this?