Ground to VB?

[SprinkleSpraycan]

I drew a schematic from the Demeter Fat Control layout and breadboarded it. Works well, does what it should. However I've never seen a ic ground pin also go to vb. I'm using a tl072. It certainly doesn't work without the 100k to vb. Can someone explain whats happening here?

[ElectricDruid]

It just looks weird because you've got the 100K/100K divider that makes the Vbias (VB) supply split into two bits.

If you join it up again by moving that 100K, it makes much more sense:

[antonis]

edit: Parallel red lines are capacitors..
(not cuts..)

[ElectricDruid]

edit: Parallel red lines are capacitors..

Why? The 1-2-3 op-amp is biased to Vb, and the feedback is taken to VB, so no need for a cap. Sure, you could add one (and I'd take the feedback to ground in that case) but there's no requirement to do so.
Similarly, the second op-amp is fed directly from the first (so the DC point doesn't change, much) and again, the feedback is taken to Vb, so no need for a cap there either. There's an argument to be made that we shouldn't be amplifying frequencies we're not interested in, so the caps are *necessary* to limit the bandwidth, but you didn't make that argument - perhaps that's what you're thinking? Could you be a bit less concise?!?

There is one part that I do think is odd, which is that 1K/1u along the top going to what looks like the volume pot. I think there's a mistake in that part and probably a missing ground connection too.

[FiveseveN]

what looks like the volume pot.

It's a mix control that pans from the flat 1st stage to the low-passed second one.



Let's just say there's room for improvement.

[antonis]

Why?

To roll-off DC gain to unity, Tom..

edit: Now realized that the circuit concerning VB is all wrong..
(didn't see your proposal above..)
So, let's make it more user-friendly..

[antonis]

As far as original design concerns..

[GibsonGM]

I find it easier to understand if you call it "AC ground"...the caps block DC, only the AC component is returning to the center of the bias voltage, to ground thru the smoothing cap...

[ElectricDruid]

what looks like the volume pot.

It's a mix control that pans from the flat 1st stage to the low-passed second one.
Let's just say there's room for improvement.

Ah! Thanks, that makes that a lot clearer!

Why?

To roll-off DC gain to unity, Tom..

Yeah, fair enough. That's what I thought you meant when I said "the caps are *necessary* to limit the bandwidth".

So, let's make it more user-friendly..

Now, that looks a *lot* better!

[SprinkleSpraycan]

Thanks for all the info. I'm gonna try it out the new way. I see my mistake on the voltage divider now...a little dyslexia and all those 100k's I was bound to spin myself into a daze.

[SprinkleSpraycan]

The 2 added caps from the inverting sides of the opamp to ground: if I understand these don't affect tone so size shouldn't make a big difference? I.e. 10n vs 100n. Is there a reason one size is preferable?

[antonis]

Actually, the DO affect the tone, as every capacitance in any circuit does..

They form HPFs with respective series resistances (100k) of 1/2π*100k*C (in farads) corner frequency..
So, 100nF result into 16Hz, 10nF into 160Hz and so on..

P.S.
The opposite (LPF) stands for 22pF and 33k & 100k respectively..

[ElectricDruid]

You mean HPF, don't you Antonis?

The roll-off equation is the same either way ( 1 / 2 Pi R C) so the cutoffs are where he said. The caps don't affect the tone *as long as* they're large enough to pass all the frequencies we care about. The open low E string on a guitar is 82Hz, iirc, so if the cutoff is down at 50Hz or so, it won't get in anyone's way. Often you see them lower than that too because it's cheap to do. One possible motivation for not going very low is that setting the cutoff at your local mains frequency gives you a -3dB reduction in mains hum (which is better than a slap in the face with a wet badger). Going lower doesn't.

[antonis]

You mean HPF, don't you Antonis?

Thanx Tom..

[antonis]

One possible motivation for not going very low is that setting the cutoff at your local mains frequency gives you a -3dB reduction in mains hum (which is better than a slap in the face with a wet badger).

But not better than a slap in the face with a dry badger..
(talking about double the mains rectified ripple..)

Conclusion: For getting completely rid of badgers, you need 10nF caps there..

[ashcat_lt]

You mean HPF, don't you Antonis?

Well, those caps DO make LPFs, but what they're filtering is the negative feedback, which makes the whole opamp circuit into a shelving high boost.  Basically, the negative feedback loop is like Bizarro world where everything is upside down and backwards. 

[SprinkleSpraycan]

You guys are rad. I'm gonna start figuring out how to understand some of that. super helpful. I'm glad you said 10n is the way to go. Just from trying stuff I did notice a 'dullness' with 100n. That must be from the -3db. I did settle on 10n.

[antonis]

You guys are rad. I'm gonna start figuring out how to understand some of that.

https://electronicscoach.com/difference-between-high-pass-and-low-pass-filter.html

Just from trying stuff I did notice a 'dullness' with 100n. That must be from the -3db. I did settle on 10n.

[duck_arse]

(which is better than a slap in the face with a wet badger)

But not better than a slap in the face with a dry badger..

we don't need no stinking badgers. if we had doves instead of badgers, I could get my cote.