Op amp providing virtual ground

[pokus]

Hey there,

just wondered why I never found (maybe there is) such a circuit where the virtual ground is provided by the output of an op amp in a pedal. Apart from the extra costs of the op amp this seems like a pretty decent option to provide a virtual ground to the circuit. The bias resistors could even be a lot smaller than in standard designs.

So is there any downside to this?


Let's say I have a dual op amp and only need one half for my audio signal circuit. Any problems using the other half for providing virtual ground to the circuit?

[anotherjim]

No problems with that. That's the way to do it as far as I know.

[ElectricDruid]

If you don't *need* to do it, what's the advantage? Most pedal designs are done to lowest-possible-cost, so if you can eliminate one op-amp, you do. If two resistors and cap does the job perfectly well, why throw an op-amp at it?

That said, we don't need to be so brutal for our own one-off designs. Still, I'd only bother if I happened to have an odd number of op-amps in the rest of the circuit and therefore had one left over - like the situation you propose.

It'll work fine, so if it makes you happy, go for it! Is it *necessary*? No, most likely not.

[marcelomd]

Hi,
Suhr Riot, a Guv'nor variant, uses 3 opamps for the effect. With 2 dual opamps there's one spare. They use it for the reference voltage.

[antonis]

BTW, that's the proper wiring for unused op-amp pins..

[FSFX]

If you want to provide a low impedance virtual ground type bias voltage and don't have a spare op amp then you can always use the TL2426 'rail splitter' IC that TI make just for that purpose. It is very easy to use and is in a 3-pin (TO-92) package so looks just like a transistor and takes up very little room on the circuit board.

https://www.ti.com/product/TLE2426

[Ben N]

If you want to provide a low impedance virtual ground type bias voltage and don't have a spare op amp then you can always use the TL2426 'rail splitter' IC that TI make just for that purpose. It is very easy to use and is in a 3-pin (TO-92) package so looks just like a transistor and takes up very little room on the circuit board.

https://www.ti.com/product/TLE2426

At $2.45 a pop, personally I'd rather use a leftover half of a dual opamp if I have one.

[FSFX]

At $2.45 a pop, personally I'd rather use a leftover half of a dual opamp if I have one.

Well, if you actually read what I said then you would see and understand that my comment referred to the situation where you didn't have a "leftover half of a dual opamp".

[CheapPedalCollector]

From what I understand the circuit does two things, it creates a high impedance ground reference, a constant current source, and allows you to use lower value resistors, which cuts down on resistor noise. It seems like a good idea particularly in high gain circuits.

[Ben N]

At $2.45 a pop, personally I'd rather use a leftover half of a dual opamp if I have one.

Well, if you actually read what I said then you would see and understand that my comment referred to the situation where you didn't have a "leftover half of a dual opamp".

Ouch. Ya got me.
But still, serious question: even if you don't have a leftover half, why would you spend this much rather than use a jellybean opamp? We're talking about stompboxes here, not NASA.
(Full disclosure: I raised this very question regarding a build of a Dead End FX Redstone board that Dino was kind enough to give me and that uses railsplitter like this. Dino said he thought a simple voltage divider ought to work, and some other folks suggested a transistor or opamp based buffered Vref as a cheaper alternative that would be good enough for rock'n'roll.)

[FiveseveN]

From what I understand the circuit does two things, it creates a high impedance ground reference

Low impedance, as to not impede the flow of current.

a constant current source

Constant voltage. CCS is very different.

and allows you to use lower value resistors, which cuts down on resistor noise

Higher value resistors, for lower current waste (does anyone use batteries anymore?). Their noise is shunted by the capacitor.

[CheapPedalCollector]

That's completely opposite from what other people have explained to me, and from what I've read about using a buffer in a voltage supply. Also in implementations I've seen.

[antonis]

That's completely opposite from what other people have explained to me, and from what I've read about using a buffer in a voltage supply.

You might got them wrong..

A real-world voltage source has low internal resistance (considered in series with its output)
//   //    //    current     //     //  high   //           //                //       in parallel  //       //
Low value resistors are needed when Vref is directly fed on a load of resistance considered significant for loading the divider output..

[FSFX]

But still, serious question: even if you don't have a leftover half, why would you spend this much rather than use a jellybean opamp?

Well, this is a quote direct from the TI datasheet of the TL2426 -

A designer need not pay the price in terms of board space for a conventional signal ground consisting of resistors, capacitors, operational amplifiers, and voltage references.
The performance and precision of the TLE2426 is available in an easy-to-use, space saving, 3-terminal LP package.

A simple TO-92 type package takes up far less space and is dead simple to wire on Veroboard or route on a PCB.

[Rob Strand]

That's completely opposite from what other people have explained to me, and from what I've read about using a buffer in a voltage supply. Also in implementations I've seen.

There's two angles to it.

When you pass an audio signal *through* a buffer the input side is high impedance and the output side is low impedance.  The input source only sees and cares about the input impedance.   The thing you also gain with the buffer is the lower output impedance, which may or may not be important to the circuit.

For a Vref rail splitter you want a low impedance.  Normally the 2x10k's and cap provide low enough *AC* impedance for the circuit to work.  For some circuits that's not good enough and you need to use an opamp rail splitter.    The important aspect here is the low output impedance.   The input side is virtually irrelevant since there is only a DC input there, which your are making up.  You can put 1k to 1MEG resistor on the input side it won't affect the output impedance which is the important aspect *in this instance*.   A value like 100k is common since it doesn't use up unnecessary current and lets you get good filtering with small-ish caps.

An opamp rail splitter is good when you have large DC currents going down Vref.   Also things like LED currents on compressors where a resistive Vref circuit is likely to cause problems.

[CheapPedalCollector]

OK I see, it's just a misunderstanding.

The input side is high impedance and I was told that this helps to reject noise from power supplies. The output impedance is low so you can use lower value resistors to VREF to cut down on noise as well. I'm not at all considering the circuit being used in an audio path, only as a VREF virtual ground on a single ended supply.

Also yes, having more stability with higher current draw.

Am I still misunderstanding?

[bartimaeus]

as ElectricDruid says, most pedals are trying to stay dirt-cheap and don't want an extra opamp. even so there are lots of pedal designs that use that circuit. but they usually aren't drive pedals, since those rarely have a noticeable current draw. certain pedals with LFOs can really benefit from the buffer Vbias though.

the Boss BD2, FT2, and TR2 use it, and i'm sure other Boss pedals too. and you can find it on some of the later, more complex DOD pedals such as the FX54.

[FiveseveN]

Am I still misunderstanding?

Looks like it. You say "you can use lower value resistors to VREF" but VREF is the output and there are no resistors on the output in the original schematic. Whole point of a buffer is to isolate the input from the output.

[CheapPedalCollector]

Yes I mean from say an opamp non inverting input to VREF, that resistor. Instead of the usual 1m or 470k it can be lower like 47k or even less.

I don't mean the voltage divider at the input, or anywhere else. The output of the opamp is the VREF rail.

[antonis]

Let's talk abour resistance and current..
(impedance should be rather confusing here..)

Below is a humble voltage divider feeding a load..
(Vth & Rth are Thevenin equivalents for voltage & resistance respectively..)



For a "stiff" Vth, meaning Vth almost equal to Vref despite RL variations, Rth has to be < 10 x RL (an arbitrary rule of thumb)
(the lower the Rth value the lower the voltage drop across Rth and the lower the Vth - Vref value - for any given load current..)

The above should stand for directly feeding a load of considerable resistive value compared to Rth with desirable Vref..
For an op-amp buffer, R1 & R2 values can lie in the lower margins of the megaOhm region without any DC offset issue..
(noise is a different kettle of fish..)

Also, see: http://www.geofex.com/circuits/biasnet.htm