Cathodyne vs. LTP question

[Lectroid]

I have an old 6V6 chassis I want to turn into a push-pull 10-12 watt amp with decent clean volume and a fair amount of overdrive as the volume is pushed up.  Somewhere here (can't find it now of course) I once saw a discussion of the merits of 2 preamp triodes and an LTPPI, a la Hoffman Stout, vs. 3 triode preamp with a cathodyne, like a Princeton AA1164.  I don't remember the consensus opinion.  Loud guitars cause memory loss.

Also, any design suggestions, especially outside the Fender/Marshal realm?  I've been reading schematics here and at Schematic Heaven for a couple of hours but nothing's jumping out at me.  An older design would be cool but I'd rather one that uses tubes we can still buy. 

Any thoughts?

[sluckey]

A big difference between cathodyne and LTP is the cathodyne has no gain, but LTP does. For that reason there is usually a gain stage just before the cathodyne. Many of us consider that gain stage to be part of the phase inverter. IOW, the princeton three triode plus cathodyne could be considered a two triode + cathodyne phase inverter with gain. If you count the number of triodes in the princeton and the Stout you'll see they have the same number.

If you want something that doesn't sound like Fender or Marshall, look at the Valco, Danelectro, Silvertone stuff. Plenty of schematics in Hoffman's library. Here's one...

     https://el34world.com/charts/Schematics/files/Silvertone/Silvertone1482.pdf

[Joel]

I recently did this:  AX84 Clean Preamp into a Fet-a-Dyne into a cathode biased PP 6V6 output section.

If you want to flirt with heresy, adding a MOSFET to perform the Cathodyne function frees up a triode to do anything else you might want; extra gain stage, one tube tremolo, one tube reverb, etc.  Hey, you could go with 3 gain stages, a one tube reverb, and a MOSFET tremolo!  Double the Heresy!!!
I added a MOSFET Tremolo to my Single stage AB763 and it works pretty well: https://el34world.com/Forum/index.php?topic=26928.msg295792#msg295792

You could also look at the other AX84 preamp building blocks for preamp inspiration.  Any of them should work if bolted onto the front of the Fet-a-Dyne.

[bmccowan]

I have an old 6V6 chassis

How old is old? Is it all octal?

An older design would be cool but I'd rather one that uses tubes we can still buy.

New production 6SL7s and 6SN7s are available from multiple sources. For pentodes, 6SJ7s are available NOS for cheap.
Adding Gibson to Sluckey's list.
I like what Sluckey came up with for Rocky. I'm building something similar in an old Knight mono Hi-Fi chassis now.

[Lectroid]

Good answers, thanks.

It's the AA964 Princeton clone I made last year sometime.  Never bonded with it although building the Trem-O-Nator was fun.  My goal is to re-cycle it into a clean platform for pedals, with a decent level of built-in overdrive possible.

My first post left out the important question: 
Given the cathodyne's not-quite unity gain vs. the LTPs possible gain of <30, is there a way to distinguish between their overdrive behavior?  IOW, given an all-other-things-being-equal amp design in the 10W range, is a) a 2-triodes plus LTP  topology, or b) a 3-triodes plus cathodyne, better for producing a warm overdrive?  Or is it a toss-up?

@bmccowan, sluckey
I'm fine with octals per se, but many of the old designs have two or more input channels, each using one-half of a triode.  I just want a single two-jack channel.  I'm still looking but so far I like the Gibson GA-16 the most:     https://el34world.com/charts/Schematics/files/Gibson/Gibson_ga16t.pdf

Bonus question: anyone know why Fender had 1.5K grid stoppers on the 6G2 Princeton but not on the AA964? Was it just that the AA964 ran at higher voltages, more chance of grid blocking?

[sluckey]

Bonus question: anyone know why Fender had 1.5K grid stoppers on the 6G2 Princeton but not on the AA964? Was it just that the AA964 ran at higher voltages, more chance of grid blocking?

Bean counters.

[bmccowan]

I'm fine with octals per se, but many of the old designs have two or more input channels, each using one-half of a triode.  I just want a single two-jack channel.  I'm still looking but so far I like the Gibson GA-16 the most:     https://el34world.com/charts/Schematics/files/Gibson/Gibson_ga16t.pdf

That is true, but there are ways to handle that, such as parallel wiring of V1. Or using the second half as a gain boost. Or just using one-half and in 10 years when the 6SL7 gets tired, rewire the socket for the other half, etc.
GA-16T - I have rebuilt/repaired many different Gibson amps, and that is one of my favorites. I finished one just recently and another forum member is building one now - active thread. Many on Fender Forums say its a clone of the tweed 5F11 Vibrolux, but its not. The Fender is fixed bias and the Gibson is cathode biased - other more minor differences too. BTW you may have found this, but there are multiple versions of the GA-18 and one of them is identical to the 16T.

[brewdude]

I have built a couple amps that use a 6SJ7 pentode gain stage before a 12AU7 (triodes wired in parallel) cathodyne PI with good results.

[Lectroid]

Bean counters.

@bmccowan,
Thanks for the GA-16 affirmation.  I'll find that thread.

@brewdude
I converted a Bell&Howell projector amp with a first stage 6SJ7.  It's a workhorse and really breaks up nicely, close to a 5E3.  It was my first amp and I still play it often.

[tubeswell]

Given the cathodyne's not-quite unity gain vs. the LTPs possible gain of <30, is there a way to distinguish between their overdrive behavior?  IOW, given an all-other-things-being-equal amp design in the 10W range, is a) a 2-triodes plus LTP  topology, or b) a 3-triodes plus cathodyne, better for producing a warm overdrive?  Or is it a toss-up?

Somewhat depends on the type of output tubes. Tubes with bigger bias voltages need more input signal to overdrive. An ideal cathodyne should be perfectly balanced (and that’s why HiFi nuts like them) and have low output impedance from both outputs, but the peak to peak output voltage from each output is limited by the available Power Supply voltage. (Some amps have an extra triode between the cathodyne output and the power tube input to bump the gain). It’s possible to get a cathodyne perfectly balanced if you make the cathodyne fixed bias, or are careful with a DC coupling set up from the previous stage.
A Long Tail Pair output can also be ‘balanced’, if you make the tail resistor large enough, and use the same size plate resistors on each output, but that will rob you of gain (because, all other things - I.e. supply voltage - being equal, the bigger you make the voltage drop across the load resistors, the smaller the plate-to-cathode voltage will be).

[Lectroid]

Tubeswell,

Excellent, you've about tripled my knowledge of PI's in one paragraph.  Thanks!