Sanity check on AA1164

[Lectroid]

My current project is an AA1164 Princeton Reverb, minus the tremolo section and using the short board Hoffman sells. 
Without tremolo, the amp no longer needed one/half of a 12AX7, and I saw that if I subbed in a 1-tube reverb, I could dispense with the other half of that 12AX7.   But while doing it, questions arose--I realize I don't understand some parts of what I've done.  (understatement:  ).

Changes from Hoffman's layout:

• Replaced the 12AT7/12AU7 Fender driver/recovery circuit with a 12DW7 one-tube reverb circuit from tubenit's gallery.
• Replaced a 1K resistor between A and B power supply nodes with a ~14H choke. 
• Added a pre-PI 100K master volume.

1. Is the 1M Dwell pot (Fender value) enough for that circuit?  Would 500K be better?

2. The 470K resistor following reverb recovery stage--what is its function?  In tubenit's version, it's a 330K.  Is it critical?

3. Can someone sketch out the bias circuit to me?  Have I re-arranged it correctly on my drawing?  I have built a Marshall-like bias supply circuit before, but this one eludes me.  Can someone explain how to calculate or estimate its range of adjustment, or just how to attack the problem?

4. Is using a pre-PI master volume a good idea for this Fender tone stack and the cathodyne PI? 

Thanks in advance!

link to file of Hoffman's layout document: https://el34world.com/Hoffman/files/Hoffman_PrincetonReverb_NoTremolo.pdf

[ac427v]

Little Details: you could change your schematic to label the reverb mix stage V3a and the pi stage V3b 12AX7.
Change the third question wording to match the schematic. 1Meg master volume. Since you asked...my preference is the LarMar post pi master volume.

Modern power transformers often result in excessively high B+. This is made worse by eliminating one tube from the Princeton Reverb circuit. You can reduce the negative effect somewhat by replacing the planned choke with the original 1000 ohm resistor. Or if necessary, a larger 2500 ohm resistor. The idea is to reduce high screen voltage on the power tubes and supply voltage to the Reverb Driver Tube.

Definitely use a 1 Meg pot for the dwell control. If you need a Dwell Control.

The Sanity Check question???
We're all Bozos on this bus

[Lectroid]

1Meg master volume...my preference is the LarMar

Badly worded question.  I was wondering about a cathodyne PI and whether one type of MV works better with it.  Or is the Lar-Mar superior, regardless of the type of phase inverter? IYO

Modern power transformers often result in excessively high B ... reduce the negative effect by replacing the choke with the original 1000 ohm resistor. Or if necessary, a larger 2500 ohm resistor. The idea is to reduce high screen voltage on the power tubes and supply voltage to the Reverb Driver Tube.

Good call, I hadn't thought about that missing tube.  But don't we want the reverb driver voltage to be as high as high as possible?  I thought that was the purpose of powering it from the B node?   Either way, your 2500 Ω dropping resistor is a good place to start.

The Sanity Check question???
We're all Bozos on this bus

Somebody has to do something, and it’s just incredibly pathetic that it has to be us. – Jerry Garcia

[tubenit]

You can use a much lower resistor value that the 3.3M/10p in the reverb circuit.  I'd suggest maybe 470k with no cap across it.  I've used as low as 220k there.  A lower value would allow more signal to go thru.


The original Fender circuit only has a 100k reverb pot and IMO, it needed the 3.3M.   However,  the one tube reverb has a 1M reverb pot with a 470k resistor from it.



And I'd stick with the 470k coming from the reverb pot.  You could even try 330k there is wanted.   

With respect, Tubenit

[mresistor]

Moving the reverb driver B+ to a lower voltage voltage PS node has been discussed here, and it is mentioned to preserve the life of the 12AT7. That tube is the hardest working tube in the amp and the very high B+ is a concern to some. Personally haven't done it yet but I don't see why lowering the B+ is going to affect the tubes operation.

[Lectroid]

You can use a much lower resistor value that the 3.3M/10p ...maybe 470k with no cap across it.  I've used as low as 220k there.  A lower value would allow more signal to go thru.


The original Fender circuit only has a 100k reverb pot and IMO, it needed the 3.3M.  pot with a 470k ...  stick with the  1M reverb pot 470k coming from the reverb pot.  You could even try 330k there...   

@tubenit, thanks for all this, including the theory behind your ideas--always appreciated!  I'll remove the 10p cap and drop the resistor to 470K

Issue?  The original Hoffman circuit wired the triodes of the 12AT7 in parallel like Fender did, with a 2.2K Ω cathode resistor split between the tubes, giving them each a cathode resistance of 1100 Ω?  Looks like 2.2K is probably too high a cathode resistance for just the one side of the 12DW7.  Or is it?

Q:  What's a good cathode resistor value there?  My instinct would be to go with 1.2K, maybe 1.5K.

@mresistor,
I shoe-horned a one-tube reverb into Hoffman's design.  It's not the12AT7-based Fender circuit he used.  I've used the B node voltage for this 1-tube circuit before, and it worked great at ~305-310 V. 

Maybe Fender runs so much more voltage in their version bc it's going thru two triodes?  Just a possibility. If B node voltage is too high, I still have that untapped C node.

[tubeswell]

4. Is using a pre-PI master volume a good idea for this Fender tone stack and the cathodyne PI? 

Use a post phase inverter MV (if you must have an MV on this). The way you have your MV drawn (with the pot replacing the cathodyne’s 1M bootstrapped grid resistor), won’t work because the input signal at the grid won’t be grounded when the pot rotation is at ‘full cut’ and it will merely mess up the cathode-bias of the cathodyne.

[Lectroid]

4. Is using a pre-PI master volume a good idea for this Fender tone stack and the cathodyne PI? 

Use a post phase inverter MV (if you must have an MV on this). The way you have your MV drawn (with the pot replacing the cathodyne’s 1M bootstrapped grid resistor), won’t work because the input signal at the grid won’t be grounded when the pot rotation is at ‘full cut’ and it will merely mess up the cathode-bias of the cathodyne.

Tubeswell,

Thanks for pointing that out.  I know I've see some designs with a single pre-PI mv so they must work in some situations.  I've done a Lar-Mar type and I can do another one.  But I'm curious if there are any 'rules of thumb' about whether to combine certain types of master volume circuits with certain phase inverter circuits? 

[ac427v]

Most 12A__ tubes have a Design Center plate maximum voltage rating of 300 or 330 volts. This is often exceeded in classic Fender designs without problems. But using those new-fangled attenuators or master volume controls result in far higher than expected signal voltage to the reverb driver tube. That does cause damage to the circuit components which are already pushed beyond recommended limits.

I have fantasized about using a dual pot with one section controlling volume and the other section reverse wired to control reverb dwell. Then the reverb input signal would automatically reduce to a safe level as the preamp volume was increased. Never tried it though.

Pre PI Master Volume controls are often used in amplifiers designed to create distortion in the preamp circuits. Dumble and some Marshalls are good examples.