Liquid Sunshine Gain Control Help

[Frank_NH]

Hi all,

I've been playing around with the Subdecay Liquid Sunshine V2 circuit using IVIark's vero layout:

Liquid Sunshine Vero

I ordered some 2N5458s and they have worked well.  I did some research and found that reducing the 100K resistor before the output volume to 10K improves the sound nicely!  The overdrive is definitely voiced with enhanced trebles, so the bass boost and treble cut options are a must for better versatility.

Now my main concern is the second gain pot (B10K).  Here's an older schematic for reference:

Schematic

This pot has a number of issues.  First, it sweeps very non-linearly from no sound to a slightly boosted sound with enhanced treble.  Since it is also the source resistor for the second jfet, I'm wondering how you should adjust the trimmer with respect to this gain pot.  But mainly, I would like to redesign this to make it more useful.  One thing I tried to get it not to turn off completely is to add a 2.2K resistor between the pot and ground.  But it still doesn't seem to do much.

Of course, even the first gain pot is tied to ground and hence can turn the sound off completely.  I think that gain controls should never go the "zero" but instead run from "clean" to max dirt.  I suppose a fixed resistor between the pot and ground would address this too.

Any ideas would be appreciated.  Thanks!  

[induction]

The first gain pot is really just an input volume control for the second transistor. If you want it not to go to zero, then yes, you can just add a resistor between lug 1 and ground.

The second gain pot does two things. Its value sets the DC gain (and therefore the DC bias in conjunction with the trimmer) and its setting controls the AC gain (the 220nF between lug 2 and ground acts as a short to ground for AC signals above a certain frequency). If you decide to add an extra resistor here, you will change both the bias (which can be adjusted back to where you want it with the trimmer) and the voicing. You might be able to adjust the voicing to your taste by changing the bypass cap value, but it will be a matter of experimentation, so I recommend doing it on the breadboard, so you can figure out what you like best before you commit to it. Swapping caps on a breadboard takes seconds, but doing it with solder on a vero takes much longer, and can ruin the build if you're not careful.

You can also change the taper of the gain pot so it adjusts more naturally. I don't know which taper to recommend because you haven't specified how it behaves. Does it reach maximum at low settings and then do nothing more as you turn it up, or does it not do much at low settings, and then turn on all at once at very high settings? Or something else?

To answer your question about adjusting the trimmer: The trimmer is used for setting the DC bias. The setting of the gain pot should have little to no effect on the DC bias, but the total value of the pot will. So if you adjust the bias with a DMM, just have gain turned all the way up. If you do it by ear, you'll want to see how it sounds at all gain settings.

There are other ways to design this gain pot, but a lot depends on what you want it to do, so you'll have to be more specific about your goals. If you just don't want it to go to zero, you may be able to just use a smaller pot and readjust the trimmer, but again, you'll have to experiment to see if that gives you what you want.

[Frank_NH]

Thanks a lot for you comments, Induction!  

I guess my goal with the second gain pot (B10K) is to have it go smoothly from some minimum level to a mid-upper boost.  In fact, here is what the controls are supposed to do according to the online description:

"Each drive knob controls a separate gain stage, and are very interactive. The top drive knob pushes the full frequency range in to a mild overdrive. The lower drive knob pushes just the middle and high frequencies, and colors the sound quite a bit. Many distinct voicings are available by simply twisting the knobs. Don't worry if you like that mid boost that so many overdrives have you can still dial it in. If you don't want it, you can dial it out. The best tones are somewhere in between."

So the 100K pot controls the overall drive, while the second knob boosts the middle/high range.

Right now, I have it all built on a vero but have been experimenting with the circuit via wires and alligator clips.  I've already tested the 100K --> 10K mod, and that works well.   Now I want to mod the second gain pot, and I don't mind desoldering as long as I know what ever I do will work reasonably well.  By the way, I also tested the bass boost, and it works but is very subtle.  May have to play a bit with that.

This circuit has a lot of potential (sounds pretty good already) and will be boxed once I finalize the mods.

By the way, the second jfet stage reminds me of the ROG deluxe fetzer valve stage:

Fetzer Valve Revisited

I was thinking of trying to emulate that.  According to the article, the gain is proportional to ratio of the drain/source resistors (Rd/Rs) where Rd > 2*Rs.  So, I'm thinking of making the source resistance go from some baseline value down to a minimum value for a desired boost level (via some arrangement of pot+fixed resistors), with the cap in there for "presence".

(Need to get a breadboard I guess...)