Bias and ground

[liagasg]

http://www.buildyourownclone.com/shredder.html

I've seen and build the above circuit.
Can somebody explain the difference between the above pots topology ?
Tone stack pots  are connected to the bias voltage.
In most pedal designs, pots are connected to ground.

Do Both topologies have the same effect? or is there a hint that i cannot understand? Somebody shed some light if you can.

[Rob Strand]

In most pedal designs, pots are connected to ground.

Do Both topologies have the same effect? or is there a hint that i cannot understand? Somebody shed some light if you can.

Both methods work.  Plenty of pedals connect the pots to Vref = Vcc/2.  Even the tube screamer does it.

The reason it works is AC signals see Vref as a ground.  In fact all power supplies look like ground to AC signals.

You can argue very fine points about the advantage of each method.   I'd say for surface mount designs with crappy X7R caps you would be better off connecting to Vref in most cases.



I should be a little clearer about your case.  One of the opamps is getting it's Vref = Vcc/2 bias through the tone control.    In cases like this you can't "just" change from Vref to ground - since we a dealing with a DC issue here not an AC issue.

[amptramp]

Anything connected to Vref can upset the bias and if there is a choice between connecting to Vref and ground, I connect to ground unless there is a good reason to need the bias voltage.  Letting AC signal currents into the Vref supply may introduce some feedback that changes transient response or could cause outright oscillation.

[FSFX]

Letting AC signal currents into the Vref supply may introduce some feedback that changes transient response or could cause outright oscillation.

I think a lot of people think having an electrolytic capacitor decoupling the Vref will make it a good AC ground but often there can be a quite low impedance source of the signal fed back from the circuit such as from low impedance tone controls or diode clipping referenced to Vref.. In some circuits this doesn't matter as the AC content on Vref appears on both the inverting and non-inverting inputs of an op amp and so is a common-mode type of signal and will cancel. In other cases the AC noise on Vref will be out of phase with the input signal and will act like negative feedback. What you don't want is that AC signal to be in phase with the signal and create positive feedback and potentially cause oscillation. So the effect on each circuit needs to be considered individually.

Here is the effect on the AC signal fed through to the Vref of a Blues Breaker for different decoupling capacitors.

[GGBB]

In some circuits (not the Shred Master but the Marshall Blues Breaker is one example), pots may be referenced to Vbias instead of ground because coupling capacitors are not used and therefore the pots carry DC as well as AC. One purpose of using Vbias in these cases is to avoid DC current flow through the pot, which can "upset the bias" (as previously mentioned) and cause pot adjustments to sound scratchy. If both sides of the pot connect to the same DC reference, there will be no DC current flow.

[Rob Strand]

Here is the effect on the AC signal fed through to the Vref of a Blues Breaker for different decoupling capacitors.

There has to be a voltage there but is it a problem?  In most cases not.

You can take undesirable voltages further.  Look at this situation, a poor ground trace,



It doesn't take much to develop 5uV.

If you are doing precision measurements then stuff like this can be as large as the signal you are measuring.  Even the wires to sensors can be a problem.   That's why test equipment tends to use differential methods (Kelvin connections etc).

Single ended set-ups are always the worst but that's what most pedals use!

[FSFX]

You can take undesirable voltages further.

There are many sources as you are well aware. The ESR of the electrolytic capacitors are just one. That is why I used models for Nichicon capacitors with the BB simulation.

[Rob Strand]

There are many sources as you are well aware. The ESR of the electrolytic capacitors are just one. That is why I used models for Nichicon capacitors with the BB simulation.

Yep, ESR can come into it and as the Vref cap dries out as it ages the ESR will rise.

[GibsonGM]

That's a great point guys. Have either of you experienced say, a repair that turned out to be noise on Vref due to ESR changes over time?

Now that it's mentioned, I can think of a few things I've worked on that had symptoms consistent with this.   

[Rob Strand]

That's a great point guys. Have either of you experienced say, a repair that turned out to be noise on Vref due to ESR changes over time?

Now that it's mentioned, I can think of a few things I've worked on that had symptoms consistent with this. 

Off hand I seem the recall a few threads with bad Vref caps on this forum.

No real magic in the ESR, as an experiment you could put a resistor in series with the Vref cap then increase it until you hear a change.   No doubt vastly different results on different pedals and different types of pedal.   Perhaps even a few still working fine with the cap completely pulled.

[antonis]

As many guys here have said: Leave Vref alone..
(can't figure out any ground more ground than original ground..)

[Rob Strand]

As many guys here have said: Leave Vref alone..
(can't figure out any ground more ground than original ground..)

Vandals get into everything these days.