Designing an Input/Gain Stage for clean to mean

[AZJimC]

I'm re-designing an input stage for my amp, that utilizes a 12a_7. The objective is a gain stage that will go from sweet and smooth to marshall(ish) hot and gnarly, all while keeping noise floor low, and having a low output impedance to drive tone stack. There will be another gain stage after tone, and into a LTPI, driving a pair of 7591A's. There is a PPIMV between the PI and PT grids.

I hope to add a relay/switch to enable gain boost switching, possibly via disconnecting the first triode cathode bypass cap. (or both bypass caps?) All values are subject to change to accomplish stated goals.

I almost have the full schematic drawn, but I hope to settle which way I am going with the input, before finishing it. Currently, the amp works great, with just one triode going to another triode, (the dual gain pot is wired before, and after the 2nd stage) then tone stack, and gain/PI, which is a paraphase type.

Here is a schematic of what I've got in mind.

Thanks for your input.
Jim

[kagliostro]

Ahem .... I'm not able to follow the signal path, please, can you explain a bit ?

pin 7 is connectet do pin 2 ?

Thanks

Franco

[AZJimC]

Ahem .... I'm not able to follow the signal path, please, can you explain a bit ?

Okay, the plates are tied together, so are the grids. for parallel triodes (note 82K plate load). The cathodes are left separate, V1a biased rather hot, and V1b biased cool, without bypass, until you turn gain up, which gradually brings in a smaller bias resistor, and a .068 cap to enhance highs in higher gain settings.

Thanks Franco,
Jim

[kagliostro]

Thanks to you Jim

So there is a junction between the 82k resistor and the .047cap (if you put there a junction dot it will be more evident, if I can say)

If I can ask ... why a ganged pot ? Or it is a dual shaft pot ?

Franco

[AZJimC]

It is ganged, and I had even thought of looking for a 5KLx250Ka pot if one could be found, or assembled from parts?  The idea is to actually increase the gain of the parallel pair, while shaping the frequency gradually toward the Marshall gain stage specs, while raising the output level.

[Colas LeGrippa]

there is no need using a relay for switching a cathode cap. a regular footswitch hooked up right onto the cathode is fine. Even 50 feet.


Colas

[AZJimC]

there is no need using a relay for switching a cathode cap. a regular footswitch hooked up right onto the cathode is fine. Even 50 feet.


Colas

Good to know, I've never tried that experiment, but that could simplify gain boost for me. I also thought that the first stage resistor could be a bit bigger to keep that triode from drowning out the tone shaping of the second triode.

[AZJimC]

A different way of achieving the same results came to me, so I revised the approach to bringing the second triode in along with gain. Still have grids tied together, but plates are separated. Each triode has its voicing, via the cathode caps, so, in effect, there would be a treb boost as gain increases.

One big plus is that I can use commonly available pots.

EDIT: attached wrong image, removed, and updated.

[jazbo8]

Not sure about the clean-to-mean part since you are just showing the first stage of the amplifier, but what you are doing is the same as the Marshall Vintage Modern or other similar models.

[AZJimC]

I don't readily find a schematic for a Vintage Modern, but this one is a little similar to a JCM with jumper on channels. The end result I am after is parallel triodes, but, I want to bring the second up as output increases. This stage will drive the TB tone stack. a tone recovery stage then goes to a LTPI. The amp is pretty mean already, but could use a little more punch from the gain control, so I'm adding a triode and using both V1 triodes where one was. Parallel is supposed to offer about 1.3x gain, lower noise, and lower output impedance. Hopefully this arrangement will accomplish those goals.

[jazbo8]

Wouldn't volume controls alter the T/S's response quite a bit without a buffer in between?

[AZJimC]

not much in this one. The stack is low loss, and no vol on it. There is a PPIMV between the PI and the 7591 grids, where the trem feeds in at the junction of the dual pots. It is the volume control. The two triodes on input combined offer low impedance like a buffer.

[jazbo8]

The two triodes on input combined offer low impedance like a buffer.

Yes, the output impedance with the parallel triodes is halved, but there are still some interactions with the T/S depending on the volume controls' settings, just not as much as the single tube version. Also, why would the gain of the parallel triodes be 1.3x?

[AZJimC]

Not sure where I read that, but, of course that depends on other factors as well, but I grabbed one quote that explains it somewhat.

-----------------------------------
As to parallel triodes:
 
 The composite tube will have this relationship to the individual triodes it is made from: the same voltage capability; twice the current capability: twice the power rating: half the plate resistance/half the internal resistance; twice the capacitance; twice the transconductance; the same mu; and twice the plate dissapation
 
 The parallel tube will have about 70% of the noise of a single section, the voltage gains is the same as a single triode (the measured gain is often higher circa 20-30%); resistance plate is so much lower, ergo there is less loss working against the plate load).
------------------------------------------

 

[jazbo8]

the voltage gains is the same as a single triode (the measured gain is often higher circa 20-30%)...
------------------------------------------

Interesting, thanks.