question about inverting blender bias

[Wounded Paw]

This is my schematic for a blender which inverts the return signal from the effects loop.  It's for a specific pedal which inverts the phase of the signal that I'm adding a blend and gain recovery stage to.  My question is the 1M resistor between VREF and the + input of the inverting opamp.  I'm thinking it shouldn't be there but I'm not sure.  The Mini-Mixer at GGG or the Splitter-Blend from ROG (phase inverter switched on) don't use the 1M resistor at the inverting opamp.  I'm not completely sure why I put it there in the first place but I might have had a reason at the time.
Any thoughts?

[slacker]

Normally for an inverting stage you would just connect the + input straight to vref or use a small resistor somewhere between 1k and 10k.

[R.G.]

A resistor converts a current to a voltage, and vice versa. The DC input impedance of the TL072 is huge, much bigger than 1M. So there is effectively no current through it. That means...

... You're right! There's no voltage across it. So it's still a the same reference voltage.

Any value from 0 ohms up to a noticeably fraction of the TL072's input resistance works just as well for the first order operation of the circuit.

At a second order, the resistor does have some thermal noise. A lower resistor is better from that standpoint.

For best DC accuracy, opamps should always have the same DC impedance leading into their inputs. That's of little consequence here, but it will bite you someday.

Best choice for that resistor is the equivalent of the DC impedances leading into the - node. For AC-only operations, that's a very small consideration.

[Wounded Paw]

that's somewhat along the lines of what I was thinking.  I was wondering why I had inserted that 1M resistor.
Thanx.