6G2-ish build with reverb

[burt_toast]

Hey all,
Here's a draft schematic of my next build, cramming reverb into a tweaked 6G2 (sort of) circuit. I changed the 56k resistors on the cathodyne to 82k, which I've read is better for pushing 6L6s/EL34s. Thoughts?
Also, I'm not sure about the bias feed, if I go with those bigger bottle power tubes. The mains and output trannies are from an old Conn organ amp, which had 2 separate (but identical) output trannies (one for 6v6, one for 6L6), one of which I used to repair a Traynor YBA1 years ago, with great success!
This amp will likely have a tube rectifier, too, which my schematic doesn't yet reflect.
Everything look plausible?
thanks,
Scott

[burt_toast]

...and I may parallel the first triode stage (ala Matchless), or can use to turn cathodyne into LTPI, since I have an unused 12ax7 half

[tubeswell]

I changed the 56k resistors on the cathodyne to 82k, which I've read is better for pushing 6L6s/EL34s. Thoughts?

The maximum amount of clean signal through a cathodyne is determined by the HT voltage (which determines the voltage across the load resistors and the plate-to-cathode voltage) i.e., HT = Vload + Vp-k. If you increase the load resistors, but don't do anything about the HT voltage, you will decrease the plate-to-cathode voltage, thus reducing the ability of the cathodyne to take input signal (from the previous driving stage) without clipping, which will cause the signal to clip sooner, which will produce unpleasant voltage-doubling distortion and/or gain-spike distortion in the cathodyne (as the input signal gets larger) - and you want to avoid that because you want the phase inverter to drive the output tubes with as-big-a-clean-signal-as-possible to get the power amp to go from Class A to Class B.

[burt_toast]

Thanks Tubeswell, I wish I understood this stuff on your level, I'm just a dumb guitar player. I will say this though, the transformers are huge, and if I remember, the mains secondary side is 375-0-375 VAC (I left my chassis at work, and can't verify this at the moment). Not Princeton trannies at all!
Seems like I oughta re-draw PI as a long tail.

[d95err]

Some notes:

The 2uF cap on the second stage cathode will shunt almost all NFB to ground. Basically the same as having no NFB. You could use the NFB circuit from the Princeton Reverb instead.

Are you sure you will have enough signal level to drive the reverb? It seems like it could be too weak with just one gain stage and attenuated by the volume control.

An idea is to use the left over triode to boost the reverb signal before the reverb driver stage.

(I’m looking at ideas for a building a 6G3 style amp with reverb, so quite similar)

[burt_toast]

Good ideas, d95err - you've given things to ponder, thanks

[burt_toast]

Oh why didn't I pay attention in High School electronics class?
Alright, this amp is turning into more of a 50 stout, now. Cathodyne becomes LTPI. I moved my reverb insertion point to AFTER V1b, does this seem okay?
I'm fairly certain (in my fuzzy-brained way) that everything here will work now, but still unsure of how my plexi-style bias for the EL34s will translate to this bias-tremolo. Any thoughts on this?

[ac427v]

I added reverb to a plexi in a similar way. The reverb was far too weak with the 3.3Meg mixing resistor. I tried many variations before settling on a 680k. Even then, the reverb is not as lush as a Fender blackplate verb. But it works for me.
edit--
--Dave is right, I explained very poorly. The 3.3M mixing resistor sends a strong signal to the reverb 12AT7 driver tube. BUT, the fender spec mixing resistor greatly diminishes the "dry" signal going to the phase inverter. The result was a lot of boing and not much guitar. This is not a problem with the AB763 circuit because fender has another triode that amplifies the mixture of the wet and dry signals. My plexi reverb lacked that extra triode. Your plan lacks it also. The common work-around is to reduce the value of the mixing resistor. I used 680k. Less signal goes to the reverb driver and and more dry signal mixes with the wet signal coming from the reverb recovery triode. The result is a pleasing balance. Your preamp has two triodes and a low loss tone circuit so I suspect you can use an even smaller mixing resistor, possibly a 220k.

[Dave]

I added reverb to a plexi in a similar way. The reverb was far too weak with the 3.3Meg mixing resistor. I tried many variations before settling on a 680k. Even then, the reverb is not as lush as a Fender blackplate verb. But it works for me.

Pretty sure you have that backwards.


Dave

[burt_toast]

Thanks ac427v, I'll definitely watch for that.

[burt_toast]

Wait, what? (Dave)

[sluckey]

I'm fairly certain (in my fuzzy-brained way) that everything here will work now

You need a cap at the input to the PI grid. Without a cap there will be dc on the reverb pot making it sound scratchy and the pot will mess up the delicate bias on the PI.

[burt_toast]

Duh, I should've caught that. Thanks Sluckey! Bias seem okay - maybe?

[tubenit]

I would encourage you to look at Sluckey's Tremonator design. Do a search on the forum.  I've installed it before on an amp and it worked fantastic!  You can get that deep Creedance Clearwater vibrato thang going on with it.


You may want to consider it for your amp design.  If you use larger tubes than 6V6's sometimes one triode may be a struggle for a good tremolo circuit.  Sluckey's design eliminates that issue, IMO.



With respect, Tubenit

[burt_toast]

Thanks tubenit, I think I get stuck in my thinking sometimes (like, "it has to be EL34s", "it has to be a bias trem" - and it doesn't). If I stay with big bottles, it sounds like Sluckey's Tremonator is a good way to go.


...back to the drawing board

[d95err]

One idea is to simply use the Princeton Reverb topology, but replace the tonestack with the ”tweed” volume/tone circuit.

Perhaps with a split anode attenuator on the second stage, like in a 6G3.

[burt_toast]

Redrawn with Marshall fixed bias, and Sluckey's Trem-O-Nator.
Same insertion point for trem/reverb, is that okay?

[Dave]

Wait, what? (Dave)

The larger the resistor, the more reverb,


Dave

[burt_toast]

Yes, thanks Dave. I plan to use a pot in that 3.3M resistor spot, to see what works best before I commit to a fixed resistor.
And, again, is that insertion point okay for both trem/rev?

[Dave]

Yes, thanks Dave. I plan to use a pot in that 3.3M resistor spot, to see what works best before I commit to a fixed resistor.
And, again, is that insertion point okay for both trem/rev?

I don't see anything wrong with it.


Dave

[PRR]

Redrawn with Marshall fixed bias, and Sluckey's Trem-O-Nator.
Same insertion point for trem/reverb, is that okay?

And yet you have drawn TR on the left end of the 3.3Meg, reverb return on the right end.

Is that how Sluckey drew it?

[sluckey]

I'm thinking that with the addition of that 50K INT pot to the already very lossy reverb mixer you will likely need another triode at the input of the PI in order to drive the power amp to its full glory.

[burt_toast]

The closer I think I'm getting, the more confused I get. I may have to step away from this and do a whole lot more research. I clearly am in the weeds here.
Thank everybody for your help,
Scott

[tubeswell]

FWIW, Leo already perfected the Princeton Reverb using 4 pre-amp tubes and 2 x 6V6s. Known good design.


If you want to run 2x 6L6s, try raising the HT voltage for the cathodyne . You’ll also need a beefier PT and 6L6 OT. It won’t sound the quite same as a BFPR.