Diodes to gnd/no diodes switch: a way to make the non-clipping position quieter?

[DC9V]

Hello DIYSB,

Consider the classic Rat mod:



On/off/on toggle allows to switch between two sets of clipping diodes, with the clipping diodes removed from the circuit in the center position.

Quite logically, the center position sounds much louder because the signal is not being clipped. So my question is: would there be a simple way to bring the volume in the middle position closer to that of the clipped signal while retaining full functionality of the switch? I know there's not much point in doing that in a Rat, but the reason I'm asking is because I'd like to implement that in an amp in place of the stock pair of 1n4148s to ground, and the volume difference is bugging me.

[ashcat_lt]

The diodes form the bottom part of a voltage divider where the ratio depends on the input level.  Replacing them with an appropriate valued resistor will make it into a static divider.  Determining the "appropriate" value is not exactly straightforward, and probably best done by ear.  One might stick a rheostat in there either just temporarily to find a good value or permanently to allow for future adjustments.

Course, that ain't gonna work on the middle of an on/off/on.  If you are trying to do three different options, you'd need an actual SP3T.

[FiveseveN]

You speak of on/off/on trinkets yet the schematic shows a SP3T (on/on/on). If that's what you have and you want the very simplest solution, you could stick a 1K resistor to ground on that open throw. Combined with the 1K before the diodes in a Rat that cuts your signal in half. The value for your situation depends on the rest of the circuit.

But for the more common on/off/on you could use a double-throw one and have the extra poles short out parts of an appropriately-sized separate resistive divider. You can also compensate for the LEDs at the petty cost of an extra resistor (also applies to the 1st case for that matter):



Warning: I pulled the resistor values straight out of my ass, experimentation is required.
PS: ashcat_lt mostly beat me to it 'cause the forum timed out 

[Rob Strand]

I had this one in mind, it uses a single switch but it doesn't compensate for the LED level being higher than the silicon diodes - that's what the extra pole in FiveseveN's solution does.  It makes the no-diode option level the same as the diodes or a little more (up to you).

[DC9V]

Thank you very much for your help... I really appreciate it! I'm drawn back to the fascinating world of DIY electronics after years of hiatus, and unfortunately I seem to have forgotten what little knowledge I had acquired.
One thing that hasn't changed in all these years is the helpfulness and patience of DIYSB forumites 

You speak of on/off/on trinkets yet the schematic shows a SP3T (on/on/on).

I whipped up that schematic in an online editor and that's the closest symbol to an on/off/on SPDT I could find. Sorry for the confusion... 
I've considered using a SP3T as it would definitely make things simpler circuit-wise. But since I'd be mounting that switch on the chassis of the amp, a SP3T slider would need a rectangular cutout, whereas a S/D/3PDT toggle would only require drilling a round hole.

I had this one in mind, it uses a single switch but it doesn't compensate for the LED level being higher than the silicon diodes

I don't think that would bother me too much ― not as much as the more drastic difference in volume when switching both sets of diodes out of the circuit.

I'm going to give all of these solutions a try and see what works best for me in this particular situation. To the breadboard!

Thanks again for the schematics and detailed explanations, ashcat_lt, FiveseveN and Rob Strand. DIYSB rules!

[Passaloutre]

I've seen in some circuits where they use the second pole of a DPDT switch to simultaneously change the resistor controlling opamp gain so that the "no-clipping" setting matches the same volume as the clipping setting(s).

Keeley did something really cool in his Fuzz Bender where the transistor bias knob simultaneously increases an opamp gain stage as you decrease the bias voltage so the output volume stays the same.

[Rob Strand]

FWIW, the Rats with the LED bias the JFET buffer at a higher voltage than the Rats with just silicon diodes.   The buffer on the silicon unit has just enough swing for the diode peak to peak output.   The LED clippers exceed that voltage so that's why they rebiased the JFETs.