Traynor amp hybrid.

[Nixon101]

I don't think that is a good idea. A better way would be to add an R-C branch off the screen node and drop some voltage with that.

To see if it's really the power amp that makes the difference, I like sluckey's suggestion to swap the pre and power amps using a jumper. Much less effort to try it out. If that does it for you, and you are just playing at home (you won't be hauling two amps around), you could arrange a convenient pre amp out/power amp in using the extension speaker jack locations. The signal path break point is in front of C7 on the YBA-1a and in front of C12 on the YSA-1. The YBA-1A also has a cathode driven Fender/Marshal type tone stack that might be a significant part of the difference too. BTW it looks like the R18 and R19 are miss-marked on the YBA-1a schematic, likely 470k not 47k.

The resistors in the long tail pair are definitely 47k. I think I read somewhere it reduces gain in the phase inverter to keep it clean, as Pete traynor designed this as a bass amp, and the normal lower power yba as a guitar amp.

As far as the rc branch you speak of, are you saying I could design that power node, somewhat similar to a bias supply how some negative voltage is dumped to ground to kind of regulate it, except I’d be dumping positive voltage to ground? If that is what you speak of, I’ve never though of that because I’ve never seen it. And I usually think, if I’ve never seen something, there must be a reason why. That actually sounds interesting since the screens are already pushed past there max voltage rating in these amps, I was also reading some stuff, and considering how I could knock that voltage down. Ive tried studying a bit on different power supplies, and rectifiers and how you can get a “half voltage” from the middle of a certain type of bridge rec. But then I also read how that wasn’t a good idea either if I remember something about the current capabilities of that “half voltage node”.

I did a lot of studying and reading every thread there is on the sss, bias supplies, power supplies etc….. but took the year off to pursue a different hobby. I should go back and read it all again for probably the 5th time, to jog my memory on a lot of stuff. I’ve come along way in learning this hobby. And I find I’ll go back and read a thread I’ve read 3 times already and still pick up, or just understand better, a little piece of knowledge. It’s like putting a puzzle together, one little piece at a time. People like me would be lost without you guys sharing your knowledge!

[martin manning]

The resistors in the long tail pair are definitely 47k. I think I read somewhere it reduces gain in the phase inverter to keep it clean, as Pete traynor designed this as a bass amp, and the normal lower power yba as a guitar amp.

Hmm. That will roll off a lot of bass, so there's a big difference between the two power amps besides the supply voltage. I'd try jumping the YSR preamp into the YBA power amp and see if you like it.

As far as the rc branch you speak of, are you saying I could design that power node, somewhat similar to a bias supply how some negative voltage is dumped to ground to kind of regulate it, except I’d be dumping positive voltage to ground? If that is what you speak of, I’ve never though of that because I’ve never seen it. And I usually think, if I’ve never seen something, there must be a reason why. That actually sounds interesting since the screens are already pushed past there max voltage rating in these amps, I was also reading some stuff, and considering how I could knock that voltage down. Ive tried studying a bit on different power supplies, and rectifiers and how you can get a “half voltage” from the middle of a certain type of bridge rec. But then I also read how that wasn’t a good idea either if I remember something about the current capabilities of that “half voltage node”.

It can be just a series resistor and a cap to ground, where the series resistor will drop some voltage due to the current flowing to the CF. You could also make a resistive divider there. Branched power supplies like that are not uncommon. IMO the screen voltage in the YBA-1a is too high.

A half voltage screen supply works fine (plenty of current) if you use a FWB with a center tapped PT secondary.

[thetragichero]

BTW it looks like the R18 and R19 are miss-marked on the YBA-1a schematic, likely 470k not 47k.

if these are the grid resistors on the phase inverter they likely are 47k (some are 47k, some 100k, and i believe some were 470k). this is one of the few tweaks to the stock circuits i do. keep in mind as a bass amp these were intended to provide a clean signal so lowering the gain from the phase inverter was intentional

[martin manning]

if these are the grid resistors on the phase inverter they likely are 47k (some are 47k, some 100k, and i believe some were 470k).

I see in the YSR they are 470k; in a typical guitar amp they are 1M, occasionally 470k. 47k creates a huge bass cut as mentioned above. At ~2-3 kHz the gain is the same as with 470k.

[thetragichero]

there are many revisions to the yba1(no a) that have various values of these resistors. the one i had had 47k. i swapped out for 470k or 1M (i don't remember which) as those were closer to 'marshall' value. i may still have the 47k carbon film i pulled out in my parts bin (i like to save stuff that tests within tolerance)

[Nixon101]

The resistors in the long tail pair are definitely 47k. I think I read somewhere it reduces gain in the phase inverter to keep it clean, as Pete traynor designed this as a bass amp, and the normal lower power yba as a guitar amp.

Hmm. That will roll off a lot of bass, so there's a big difference between the two power amps besides the supply voltage. I'd try jumping the YSR preamp into the YBA power amp and see if you like it.

As far as the rc branch you speak of, are you saying I could design that power node, somewhat similar to a bias supply how some negative voltage is dumped to ground to kind of regulate it, except I’d be dumping positive voltage to ground? If that is what you speak of, I’ve never though of that because I’ve never seen it. And I usually think, if I’ve never seen something, there must be a reason why. That actually sounds interesting since the screens are already pushed past there max voltage rating in these amps, I was also reading some stuff, and considering how I could knock that voltage down. Ive tried studying a bit on different power supplies, and rectifiers and how you can get a “half voltage” from the middle of a certain type of bridge rec. But then I also read how that wasn’t a good idea either if I remember something about the current capabilities of that “half voltage node”.

It can be just a series resistor and a cap to ground, where the series resistor will drop some voltage due to the current flowing to the CF. You could also make a resistive divider there. Branched power supplies like that are not uncommon. IMO the screen voltage in the YBA-1a is too high.

A half voltage screen supply works fine (plenty of current) if you use a FWB with a center tapped PT secondary.

Are you saying just a simple voltage divider here (added in green) I could safely knock like…. 150 - 200 volts off? Keep in mind the screen resistors still further down the line. I’m not sure I’ve ever seen that done before, mind you I haven’t looked at every schematic for every amp before. Also the 470R resistor between the plate and screen nodes is what I think you mean by series resistance? I didn’t think you could raise that value high enough to knock that much voltage off. Wouldn’t I need a ridiculously high wattage resistor to perform that without burning up?

[martin manning]

I thought we were talking about a supply for the cathode follower driver mentioned above, which only needs 2-3 mA.

[Nixon101]

I thought we were talking about a supply for the cathode follower driver mentioned above, which only needs 2-3 mA.

Maybe I misunderstood your comment when you said, a better idea would be to run it off of the screen node with an r-c branch knocking some volts off. I may be confused as we have a few different ideas all floating around in one thread here. I think the sss runs that driver tubes plates off the screen node too, so I was thinking of killing two birds with one stone, lowering the screen volts and keeping the driver tube plate volts down at the same time. Atleast I thought that’s where this conversation/idea was heading.

Edit, I suppose I didn’t draw a green line branching off to the driver tube plates. Also was under the assumption the negative voltage for the driver tube was derived in a similar fashion to the sss.

[Nixon101]

Also I wanted to say, I don’t want you to think I’m being reluctant to try your bipolar supply. I’m just trying to cover all my options as with the stock ysr-1 iron has a dedicated bias winding, while the “transplant” iron doesn’t. Also the new iron and the power/bias supply would be more similar to the sss. I tend to understand classic circuits. Fenders, Marshall’s, traynors. Anything too complicated, or that stray from that topology, I get a little lost. It’s just what I’m familiar with.

[Nixon101]

So after reading all your replies, and going back and re-reading the sss/post phase inverter threads, I think I’ve decided to skip the transformer transplant, and just add the driver tube set up, using the bipolar power supply method off the 60v winding . My next question would be…. If I decided to keep the tremolo, whether I add a tube or sacrifice another tube for the driver, how would this phase inverter driver tube’s bias requirements interact with the tremolo, being a bias shifting tremolo? Would I just run the new bias through the intensity pot like normal (schematic posted in first post) or would I have to separate the cathode/power tube grid bias supply from the grid of the driver tubes bias supply? Is this uncharted territory? Opinions?

[martin manning]

...I’ve decided to skip the transformer transplant, and just add the driver tube set up, using the bipolar power supply method off the 60v winding . My next question would be…. If I decided to keep the tremolo, whether I add a tube or sacrifice another tube for the driver, how would this phase inverter driver tube’s bias requirements interact with the tremolo, being a bias shifting tremolo? Would I just run the new bias through the intensity pot like normal (schematic posted in first post) or would I have to separate the cathode/power tube grid bias supply from the grid of the driver tubes bias supply? Is this uncharted territory?

I think a good way to experiment with this idea would be to build a small board with the bipolar supply and the CF driver circuit, but replace the 12AX7 triodes with two LND150 MOSFET's as Source Followers. They will directly replace the 12AX7 with no need for a filament supply, and they cost less than a dollar. The modifications to your amp would be minimal, too. I don't see why you couldn't use the existing tremolo oscillator to modulate the power tube bias by coupling it to the SF driver bias in the same way as it was done in the original.

[Nixon101]

I’ll take that into consideration. I’m still not 100% sure what the plan is going forward, but I’ll let all this information brew for a bit and decide when the time comes. This isn’t something I’m tearing into right away but just planning for a winter project. I really hate starting something that I don’t know for sure what to do or how to do it, which is why I do a lot of research and ask a lot of questions before any work happens at all. If I come up with anymore questions in this pre-planning phase, I’ll post in this thread. If, or when I do start working on it, I may start a new thread, as this one kind of started mostly on one idea and landed on another. If anyone still wants to chime in with ideas and information, I’ll still check in even though I may be silent. And again, all help is appreciated. I think I mentioned before that I had taken the summer off to pursue another hobby and am just getting back into this….. I had went looking for my breadboard and found still rigged up from last winter when I was experimenting your bipolar power supply voltage multiplier. So technically this is actually year two of planning this project lol. I remember thinking I didn’t want to impulsively alter this vintage amp last year, but if I still wanted to months down the road then it must be destined for it.

[martin manning]

I had went looking for my breadboard and found still rigged up from last winter when I was experimenting your bipolar power supply voltage multiplier.

You're half-way there then. Add the MOSFETS and resistors (6 parts) to your breadboard and connect the drivers between the PI output and the power tube grids. You can leave the existing bias and tremolo set-up exactly as it is. Run a lead to the bipolar bias supply from your 60 VAC bias winding, and a ground lead back to the amp chassis. You will have to tweak the bias circuit to get the power tubes biased where they were originally, but if you have a bias adjustment pot installed that will be easy. This way you can see if adding a driver circuit takes you in the direction you want to go before making any irreversible modifications.

[Nixon101]

Awesome information. Thank you for the schematic. So what I’m seeing is, and maybe asking for clarification, is that unlike the tube version of the driver, where the bias voltage is injected onto the cathode of the tube and on the to grids of the 6ca7’s, lets say about -45 volts….. instead is injected before the mosfet, and comes out the mosfet about -45 volts on the other side? Even though the “cathode” if that’s what it’s called, is hit with -170? I’ll have to admit I know almost zero about solid state electronics, but this doesn’t seem too complicated. Also with the existing bias supply still intact, I’m also tapping off the 60v winding, I guess in parallel you might say, to create the bipolar supply? Or is that just for experimental purposes until I actually build it in to the amp?

[martin manning]

Basically that's it. A depletion mode MOSFET Source Follower acts very much like a tube cathode follower. The output comes from the Source terminal at the top of the 100k, and it should be within about a volt of the bias voltage you feed in. And yes, to power the bipolar supply just solder a wire to the 60VAC winding where it connects to the existing bias circuit.