"Brown" knob

[stickjam]

The Hammond hot-rod project continues...

I've abandoned the idea of replacing the solid state Leslie 770 amp with a 122-style design for a number of reasons, so I'm turning my attention to adding the grit within the organ itself.

At the best point to insert this "color" stage, it scopes out at a nominal 1.1Vpp unbalanced signal. The new circuit's output will be seeing a 20K impedance. I hope to model the same balanced topology of the Leslie 122 amp in a lower-power line version. I am envisioning the signal chain of this stage being:

Attenuation -> DRV134 to balance -> 2-stage balanced tube amp based on the 122 -> iron to unbalance -> Attenuation

One design objective I have is to implement the user interface not as separate drive and output controls but as a single "Brown" knob. (I'm actually considering being cute and using a decorative knob from a toaster that has different shades of "doneness" painted onto it. :green: ) I'd like this stage to be a psychoacoustically-if-not-electrically unity gain device. With all else equal, the apparent volume shouldn't change when you turn the knob--just the color should change.

My questions are:

1. Am I going to need to add some more gain on the drive side?
2. I originally planned to use a dual-gang rotary switch and hand-selected resistors as the "brown" knob. I'd ideally like to make it more of a smooth performance control rather than having discrete steps. Is there any way a dual ganged pot could be configured to provide the intended behavior? Will I need some exotic forward+reverse combo taper thing?
3. Any suggestions on what tube type (lower power pentode) and transformer model I should use to get a line-level output that imparts color most like the 122's 6550 power amp stage?

I'm getting close to completion on the power supply (hoping 300V 30mA B+ is sufficient). I'm looking for ideas as I prepare my shopping list for my next visit to the local "toob shop" for supplies to breadboard this piece.

Thanks!

Bob

 

[mikep]

hello

what you need is a cross-fader to blend dry signal with the brown. you can use a dual linear-taper pot with appropriate "slugging" resistors from wipers to ground to get reasonably flat summed gain. I recommend voltage summing to combine the two paths. the summing resistors can be slected to also do the slugging. I built something like this with a 50k linear dual and 15k summing resistors.

mike p

 

[SSLtech]

Search my "crush-n-blend" schematic.

Keith

 

[stickjam]

Thanks -- I was actually looking for something like the "crush-n-blend" for another application. That will certainly fit the bill there. :thumb:

However I'm not sure a cross-fader is going to give me the type of effect I'm looking for. What I'm after is a tube drive control that is inversely coupled to the output gain control to maintain pretty much the same net volume level through its entire rotation. Like grabbing the Drive and Output controls on a git amp and simultaneously turning them in opposite directions--except with one hand.

The snubber resistor idea got me thinking... Would something like this get me where I want to go? I'm not sure where the bottoms of the snubbers should terminate relative to the minimum trimmers.

(The other answer I'm eagerly awaiting from those that know more about how different tubes sound is what the most likely small-signal pentodes I should try to get a similar overdrive in a PP line-out situation as a pair of 6550 beam pentodes will in a power amp. First that comes to mind is 6AQ5A. Anything else I should try?

 

[mikep]

well, that is an idea. thing is, if the amp is saturating it will get QUIETER on you with the ganged pre-gain post-atten. why fight it, the time tested combo of pre and post gain knobs is pretty good in this kind of application. the crossfader would be if you were going for a more subtle effect.

 

[stickjam]

[quote author="mikep"]well, that is an idea. thing is, if the amp is saturating it will get QUIETER on you with the ganged pre-gain post-atten. why fight it, the time tested combo of pre and post gain knobs is pretty good in this kind of application. the crossfader would be if you were going for a more subtle effect.[/quote]

I must confess that I pulled a bit of a reductio ad absurdum trick, hoping to bring out ideas to solve the problem that the output control couldn't be a direct inverse of the input attenuation.

The whole idea is to produce the dual gain control effect with an automatically adjusting output gain that psychoacoustically compensates for changes in a single input gain knob. Originally I conceived this as a simple 2P6T rotary switch to form two attenuators with resistors selected to taste.

The "one knob" design decision was initially one of practicality - to put in the place of an existing control that I am eliminating, maintaining aesthetics. I later thought it'd be cool to make this a smoother performance control, however there is not enough clearance behind the panel for a 22-step Alps attenuator.

I just thought I'd throw it out as an exercise to your collective geniuses to brainstorm the question: Is any way one can press a dual-gang pot into service to accomplish this goal? While precision is ideal, a solution that comes reasonably close is fine too. This is a musical instrument after all, not a laboratory device. :wink:

I'm a couple weeks off from being ready to breadboard this thing, so I'm finishing off my shopping list for it.

--Bob

 

[Mendelt]

Even rotary switches with selected resistors might not work. When the imput signal is lower to start with the amp will start clipping later than with a louder input signal so the input/output attenuation ratio will depend on the signal.

 

[SSLtech]

Okay, I get it.

Dual linear pot (for a mono application).

Wire it as a log gain stage... an inverting op-amp stage in other words, with the wiper returning to the inverting input, and the non-inverting input grounded.

Let me see if I can dig an example up:

Ah... here you go. Lok at the SECOND stage in this schematic:

R5 and R6 are max/min gain setting resistors. VR1 is a nice log (evenly-spaced dB scale...-nice!) control. 10k for R's 5 and 6 together with a 100K VR1 should give you ±20dB of range or thereabouts.

Build the circuit twice, BUT wire the second pot the opposite way around, so that when the first stage gives you 12dB boost, the second gives you 12dB attenuation...

I think that's what you're looking for, maybe?

Keith