Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: firemedic on November 14, 2010, 04:39:31 pm
-
I'm trying to free up a preamp stage. I have a paraphase inverter driver now- I'd like to do a split-load driver. I have more gain than I need right now so that won't be an issue. I'd also like a NFB loop with presence control. Any suggestions? The power section has 2-6L6 GCs, and the preamp has 2 gain stages.
-
With no other info, you have competing requirements.
You have a paraphase (2 triodes), and want a free triode. Split-load looks like only 1 triode, but when there is a feedback loop, it is always connected to the triode ahead of the split-load inverter. Since the split-load has no gain on its own, it always has what I consider a "pre-gain stage" in the triode right before it.
The paraphase doesn't have the full gain of both triodes. Generally, the first triode makes full gain, then the output signal is knocked down by the amount of gain of the second triode. End result is 2 signals, opposite phase, and as-big as they would be if amplified by 1 triode.
A split-load with a pre-gain stage will arrive at essentially the same-gain as a paraphase inverter.
You also want to add a NFB loop. Negative feedback works by throwing away excess gain to get less distortion. The presence control reduces the amount of throw-away of treble sounds to get the effect of a treble boost. So you need more gain than you have now to have enough to add feedback and still be able to drive the output tubes as easily as before.
By the way, the long-tail inverter will likely have roughly same-gain as the other 2 setups. From input to 1 output only, you get about half the gain you'd expect of that type of triode.
But I can't say for sure which inverter will work for you. It's more complicated than that, and depends on how big of a drive signal the output tubes need and how much supply voltage you have available. We need a schematic.
The best way to get your extra triode will be from an unused effect, i.e., tremolo or reverb. If you don't have those handy, there may need to be major surgery to truly add gain or negative feedback.
-
Okay, well the reason I want the extra stage is to recover from a tone stack. Right now my one preamp tube goes before & after the tone stack, to the PI. I want to try parallel triode for the first stage, using the other "freed up" (I know it's not really free) PI half to recover gain post tone stack, so at least theoretically I could make this work....
At this time I have TONS of gain with just the 2 stages. With a 12AY7. A net loss of gain would be okay.
I guess what I'm asking is this: does an existing circuit fit this scenario better than others? Super, Princeton, etc.? Because obviously I lack the math skills to make one from scratch.
-
It would really help if you could post your schematic. Do you have 4 tubes to work with? 2 small ones and 2 big ones? (Not counting a rectifier.) If so, read on...
Take a look at the Fender Harvard 5F10.
http://www.el34world.com/charts/Schematics/files/fender/harvard_5f10_schem.pdf
Replace the 6AT6 with a 12AX7 and parallel the two triodes. Looks like it has everything you want except presence. You would have a parallel preamp, tonestack, post amp, split load PI, and P/P PA.
-
Actually I already have posted the schem., it's the Dan-O-Lux (don't laugh) on the Schematics board. I'm very happy with it at low volumes up to 5 or 6, but when it's turned up all the way it's very muddy. It's got 2 6L6GCs, 4 preamp tubes and a GZ37(!) rectifier.
The input is more or less an AB763 normal channel. Input to 12AY7, tone stack, back to 12AY7, to tubenit's one tube reverb, to 5751 paraphase inverter. It's a 5E9 Tremolux from the PI onward. Sounds great but at higher (stage) volume it is not only muddy but the highs are pretty harsh. JAN tubes are handling the signal & they help a little with the harshness.
Since I'm going back in anyway I'm just going to start over with a new board, different caps, carbon film resistors (those CCs are noisy) and try varying the power tube bias for the tremolo while I'm at it. So I'm looking for a good split load driver circuit, so I can increase gain if need be to the PI if that makes any sense. I need to convert to adjustable bias, I know. And I don't want to give up the reverb or tremolo. AND I would like to try the parallel triode.
In short, I love my amp so much I'm going to completely destroy & rebuild it.
That 5f10 looks like it will fit the bill unless I'm missing something.
The Harvard looks like a good start, thanks sluckey. And thanks to everybody who has offered their good advice, none of this would have been possible without you.
-
Hopefully, you could also capture that .sch schematic as a .gif and post it here. Unfortunately, I don't have the .sch program on my computer anymore.