Push/pull 6V6 amp build dilemma

[AmberB]

So, I'm currently working on building a small PP6V6 amp using the transformers from the tiny chassis of what I think was a portable record player in it's first life.  The power transformer does not have a 5 volt winding so I know it didn't have a 5 volt tube rectifier.  To be honest, I don't remember if it originally had a 6 volt tube rectifier or just diodes, but I'm using a pair of diodes for that.
Anyway, I was in the process of building a sort of clone of an AA-964 Princeton, with the addition of a midrange control in the tone stack.  I have 2 9 pin preamp sockets and the 2 octal sockets. 
About 2/3rds of the way through the build, I realized that the AA-964 only has 3 triodes in the audio path.  The fourth triode is used for the tremolo circuit.  I don't have the space in the chassis for the 2 trem pots, and I wasn't going to put tremolo in the amp anyway.
So my current dilemma is, do I use the 4th triode ias a cathode follower preamp output, or do I just change out the paraphase inverter for a long tail inverter, like the Deluxe...
Was going to post pictures, but they're too big, gotta shrink them...

[AmberB]

Top side

[AmberB]

bottom side
This was with most of the parts soldered in before I realized I had 1 too many triodes in the circuits
I don't want this amp to be a high gain amp, otherwise I'd use that 4th triode for that.

[PRR]

> do I use the 4th triode ias a cathode follower

I bought a car with 4 seats. I may never have 4 humans aboard. I don't feel I have to wrangle butts to fill "all the seats".

Car seats are a couple hundred bucks each. Idle triode is, what? $10-$20?  I could stand the "waste".

[glass54]

Hi Amber
You could double up on input triode?
Or do the LTPI (as in deluxe, which I like but some prefer the Paraphase when driven hard)
Either way, no waste, just opportunities 
Nice project
Kind regards
Mirek

[mresistor]

yes  have the input triode paralleled with the unused triode, and put it on a switch..

[AlNewman]

Is it a cathodyne PI?  You could use an LTP.

[Joel]

Check out this schematic from the old AX84 core series.  I built the "added gain" version for my Woke Heretic amp (married to a solid state cathodyne, then 5E3 power amp).  Don't worry about the "added gain" name, its designed to drive 6V6 better not to distort the output section (the notes in the schematic explain).

The reason I post these for your consideration is due to the interesting parallel input stage.  You'll note that it's not simply a parallel stage but more like two different input stages parallelled.  I have mine hardwired to "blues" which can be plenty clean too, and I use the Sparkle/Grit control to dial in preamp dirtiness.  It's a brilliant preamp, I like it a lot.

Note: parallel input stage is DC coupled to the next stage, so some experimentation may be required to get it to work in your implementation.

[AmberB]

I decided to go with the Deluxe AB-763 type LTP, it just seemed to be the best idea.
It will be interesting to see how that LTP works with a cathode biased PP 6V6 pair.

[tubenit]

It will be interesting to see how that LTP works with a cathode biased PP 6V6 pair.

It has paired well on all of my amps where I've done that.   


With respect, Tubenit

[tubenit]

Here is an idea for a Minimalist 6V6 amp for you OR someone else to consider when only having two dual triodes for a preamp. 

You could leave off the PPIMV and place the FX send pot on the back of the amp.

I've used 3 different delay pedals, 2 different reverb pedals and an inexpensive phaser in this "psuedo active FX loop" and they have worked well for me.

This cheapo Kmise delay pedal is one of the 3 delay pedals I own and I like it.


Amazon.com: Kmise Analog Delay Electric Guitar Effects Pedal Mini Single Type DC 9V True Bypass : Musical Instruments


with respect, Tubenit

[dwinstonwood]

It will be interesting to see how that LTP works with a cathode biased PP 6V6 pair.

AmberB, it works great. I built a cathode biased 6V6 amp with three preamp triodes and an LTP  a while back, and I love it.

It based on the Valve Wizard's medium gain preamp. It has the tone stack at the end with a gain and a volume control. I scaled up the tone stack values, used an artificial CT for the 6.3V grounded at the power tube cathodes, and added a presence control.

Here's my schematic with voltages.

[AmberB]

dwinstonwood, interesting schematic!  I'm kind of curious why they didn't use a cathode follower into the tonestack, but I assume there was a reason for that.
Knowing how I am with unused triodes, I'd have to either use a single triode tube such as a 6AV6 or a 6CN7 instead of 1 of the 12AX7s, or find a use for the extra triode from the 12AX7...

Tubenit, that would certainly be another good way to build a small amp.

[AmberB]

I believe I have all the soldering done on this amp project.  The wiring is messy, but then, it's point to point, that's always somewhat messier than a circuit board amp.
If I build another amp using this same Hammond chassis, I'll have a better idea of how to lay out some of the parts.

Now on to the testing phase...

[AmberB]

It's basically a single channel, no trem, no reverb, Deluxe clone with added midrange control, using the AB-763 parts values.  The power supply is "improved", using a 47uf, 500v primary filter cap, a 47uf, 450v screen cap, and an F&T dual 33uf, 450 cap for the LTP and the preamp.
I used those values because I had them.  The 33uf values could just as easily been 20uf...I have a J/J 40, 20, 20, 20 cap can to use, but I didn't have a good place to put it on this small chassis.
The chassis is a Hammond 13.5 X 5 X 2 inch chassis.

[AmberB]

I did a bit of testing on the amp...first the basic voltage test, all the voltages seem reasonable with the tubes installed.  The B+ off the diodes is about 330 volts, so a reasonable voltage for 6V6s.  The cathode voltage is about 19.8 volts with the 250 ohm cathode resistor.  I'll have to calculate if that's a good bias voltage.

My first tests were with no speaker plugged in.  I plugged in a speaker and turned the amp on, as soon as it started to warm up, the speaker started making a fairly loud singing noise that went up in frequency as the tubes got hot.  I remembered something I read on the forum, so I shut off the amp and switched the power tube grid wires and tested it again, the noise was gone.  That's the first time I've had that happen.  I've never had it happen with a split load inverter, so it must be a quirk of the LTP and how the feedback loop is fed into the tail of the LTP, interesting...

I plugged a guitar in and tested the amp to see if I got sound, and it works, but I have a weird problem with the tone controls.  With them all down, the amp works.  If I turn them all up to 10, what I get is a lot of treble, and if I turn down only the treble, it's like turning the volume down.  I figure I probably have something wired wrong in the tone stack, now I have to figure that out.

[sluckey]

I plugged a guitar in and tested the amp to see if I got sound, and it works, but I have a weird problem with the tone controls.  With them all down, the amp works.  If I turn them all up to 10, what I get is a lot of treble, and if I turn down only the treble, it's like turning the volume down.  I figure I probably have something wired wrong in the tone stack, now I have to figure that out.

The tone stack is not wired correctly. Sounds like the bass and mid pots are wired backwards. An AB763 TMB tonestack operates like this...

Turn all pots to zero and you get no sound. Turn up only the treble pot and only the treble sound comes through. Kinda like a volume pot but only for the treble.

Now turn all pots to zero but this time only turn up the mid pot and only the mid sound comes through.

Now turn all pots to zero but this time only turn up the bass pot and only the bass sound comes through.

[AmberB]

In the past builds, I've mostly used Ampeg type tone circuits and split load inverters because I've been modifying Bogan amps, and that's what they use.  Looking at the Fender schematics is a bit confusing the way they draw their tone stacks,  the schematic is confusing on how the pots are wired.  I finally looked at a layout diagram for the AA-270 Twin reverb with the treble-mid-bass setup to see how the wires are hooked up to the pots.  That makes it easier to figure out how they did it.  Marshall seems to have a different way of wiring their tone stack, but they use different pot values and a cathode follower to drive the tone stack, and I don't know if that has an effect on the way they're wired.

[sluckey]

The Marshall TMB T/S is slightly different from the Fender TMB T/S. You have a Fender T/S. Don't muddy the waters by looking at a Marshall T/S.

[tubeswell]

FWIW single channel PP fixed bias amp with a James tone stack (You could use 6V6s with appropriate OT impedance ratio change).


With the tone stack in this configuration, both tone control pots at mid-rotation, gives almost dead flat Freq response. with both pots at max rotation, there is a 'mid-scoop'. With both at min rotation, there is a 'mid-hump'. (assumes the TS raw switch isn't lifting the TS out of the load and the 'slope' switch isn't shorting out the 100k)

[AlNewman]

From what I can see, the only difference between a marshall and fender tone stack, besides the capacitor and resistance values, is that the mid pot is wired as a variable resistor (fender) vs a voltage divider (marshall).  My own personal preference, I would choose Fender stack.

[AmberB]

From what I can see, the only difference between a marshall and fender tone stack, besides the capacitor and resistance values, is that the mid pot is wired as a variable resistor (fender) vs a voltage divider (marshall).  My own personal preference, I would choose Fender stack.

Yeah, the original Marshall tone stack is basically the same as the Fender 5F6A bassman tone stack.  Most of the later Marshall tone stacks that I've seen are just an expansion of that.

Tubeswell, I've actually done most of my builds from Bogan chassis using the James tone circuit.  I'm usually used the split load inverter on those amps because the Bogan amps were already configured that way.  This little amp is a bit of a departure from what I normally build.

[AmberB]

It turns out that the problem I had with the tone stack on this little amp was the way I wired it.  After wiring it the way it's shown on a Fender AA-270 amp layout diagram, this amp works the way it's supposed to.  Now I just need to figure out how to lower the gain from the preamp to the LTP so that it doesn't get so loud and distorted so quickly as you turn up the volume.  That shouldn't be too difficult, I know of a couple different ways to do that.

I have this big book called "The Tube Amp Book" by Aspen Pittman that I've had for about 20 years, the second half of the book is all schematics from a lot of the different guitar amp manufacturers.  It's not complete, but it's got close to 200 schematics in it, so it's really handy when you want to look something up.  The book I have is copyrite 2003, but apparently the first edition came out in 1986

[tubenit]

Now I just need to figure out how to lower the gain from the preamp to the LTP

Try a 12AY7 in V1.   IF you have 22uf cathode caps on the first two gain stages, try 5uf for V1a then 2.2uf for V1b.


IF that doesn't clean things up enough for you, try a 12AT7 for the LTPI or even a 12AU7.


With respect, Tubenit

[AmberB]

Now I just need to figure out how to lower the gain from the preamp to the LTP

Try a 12AY7 in V1.   IF you have 22uf cathode caps on the first two gain stages, try 5uf for V1a then 2.2uf for V1b.


IF that doesn't clean things up enough for you, try a 12AT7 for the LTPI or even a 12AU7.


With respect, Tubenit

More good ideas!  I think I have a 12AY7 somewhere, I know I have 12AT7s and 12AU7s in my stash.

[shooter]

besides what was already stated, I like rolling the AU7 in "V1's" position, if it gets smiles, i'll typically re-bias the circuit for an AU
also a scope on the signal path typically show's you where "too much gain" or distortion comes from.

[AlNewman]

Do you have a schematic of your build?

[AmberB]

Do you have a schematic of your build?

No, not really.  I just followed a couple of Fender schematics for the various sections of the amp.  I think it could best be described as the normal channel of the AA-270 Twin Reverb, without the bright switch, into the LTP and power tube section of the AB-763 Deluxe Reverb.  No reverb or vibrato involved in the amp.
The amp only has a B+ of 330 volts due to the PT from the donor amp that I used, but it's enough for the purpose.

[AlNewman]

I agree with Tubenit, especially regarding the 12AT7 in the LTP.  You could also try changing the gain structure of the phase inverter.  Or you could try changing it to fixed bias at the power section? 

[dwinstonwood]

You can also lower gain by adding two resistors in the form of a voltage divider.

I used an 820K and a 220K in this schematic. You can use this calculator to try different values, so you can choose what percentage of the voltage you want to send to ground:

https://www.digikey.com/en/resources/conversion-calculators/conversion-calculator-voltage-divider

[dwinstonwood]

AmberB, these are interesting reads about how Mike Soldano used voltage dividers to help shape and control gain:

https://www.ampbooks.com/mobile/classic-circuits/soldano-slo/

https://robrobinette.com/How_the_Saldano_SLO-100_Works.htm#Attenuator_Circuit

[AmberB]

Putting a 12AT7 in both preamp and LTP tube positions brought the amp into a bit more control, so I know I have too much gain going on in this little amp.  If you like high gain amps you would probably like this the way it is.
I don't seem to have a 12AY7 tube to try, so I have to order 1 or 2 to have around for this kind of stuff.
Anyway, now I need to decide if I want to approach the gain problem by removing the cathode bypass caps on the preamp triodes and see what happens, or just use voltage dividers in the signal line.
The preamp is a typical Fender type, but it seems extra strong as compared to my old Princeton amp, it distorts pretty early on the dial.  Also, it gets a lot louder after distortion starts...I'll keep pecking away at it...
I probably should try a 12AU7 in the preamp socket, I have several of those tubes...

[AmberB]

Putting the 12AU7 tube into the preamp socket cleaned up the amp quite a bit, so now I just need to find the best way to lower the gain on the preamp.  I don't know if I need to lower the gain on both triodes or just the second triode.  I was thinking about removing the cathode bypass cap on the second triode and see how that works.  I'm guessing that the second triode in the circuit puts too much signal into the LTP and causes early onset of distortion.
I'm not sure how the tone response of the triode will be affected by removing the cathode bypass cap.

I'd like to be able to use a 12AX7 for the preamp tube if I can.  I'm going to leave the 12AT7 in the LTP socket.

[shooter]

find a datasheet for AX7 that has one of these, sorta woks when a scope isn't handy

[AmberB]

find a datasheet for AX7 that has one of these, sorta woks when a scope isn't handy

Yes, my RCA tube manual has those charts for the various tube types, they're pretty helpful.  The one thing they don't tell you is what happens if you don't have a cathode bypass cap in the circuit.  I know that reduces the gain, but it's hard to say by how much.

So, I removed the cathode bypass cap from the circuit for the second preamp triode because I figured that was the triode producing all the extra gain that I wanted to get rid of.  That worked to some extent, but I wanted a bit more gain reduction to the LTP, so I put a voltage divider in the feed to the LPT.  I tested the amp and I had almost no signal being amplified.  It turns out that when you put a voltage divider in the line to the input grid of the LTP, you have to isolate it from both sides with a bypass cap.  Having never built an amp like this before, and not building from a specific schematic, I was not aware of that.
The amp works a lot more to my liking now.  I used a 220k resistor in series and a 470k resistor to ground.  I'm sure that could be tweaked for more or less gain to the LTP, but at some point ya gotta say "good enough".
This amp has been an education for me...
This does make me wonder if you actually need the second triode in the preamp, or if you could just feed the output of the tone stack and volume control directly to the LTP like the Princeton amp does with their split phase inverter setup...

[sluckey]

It turns out that when you put a voltage divider in the line to the input grid of the LTP, you have to isolate it from both sides with a bypass cap.  Having never built an amp like this before, and not building from a specific schematic, I was not aware of that.

And now you know. All part of the education. Probably gonna stick with you longer than if someone had just told you. 

BTW that should say coupling cap, not bypass cap.   

[PRR]

what happens if you don't have a cathode bypass cap in the circuit.  I know that reduces the gain, but it's hard to say by how much.

"About half".

Not exactly; but we rarely have a precise specification on g-amp gain.

Half, because for a wide range of conditions, the internal dynamic impedance of the cathode is similar to the resistor needed for a happy bias, because the goal of bias is to set the tube about halfway between starvation and saturation.