Capacitors in Parallel with a Vref resistor... What The Heck?

[jubal81]

A more specific question about the parallel resistor and capacitor at the (+) input of the op amp, as shown:



I understand R4 is providing a bias voltage to the opamp. But is the C17 capacitor interacting with the resistor at all in the audio spectrum, or is it independent? Based on what R.G. said about potentially coupling Vref noise from elsewhere, I'm wondering if it'd be better for the capacitor to go to ground instead, if it's just acting as a lowpass filter. But if R4 is creating a shelving response along with C17 as Tom said then that perhaps wouldn't be a good idea and it should stay as-is.

Yes, those caps should go to ground instead. For effective filtering, they need to dump frequencies to a low-impedance point and ground is best. Attaching to the VR actually allows for noise from VR to be coupled to the opamp's non-inverting input.

I don't think tying those caps to the VR was a design decision. If the circuit used a split supply, those VR would be ground. Looks to me like the designer took a split-supply design and just replaced the GND with VR on the schematic.

[aion]

So just to make sure I'm getting it, two different options:



Are these any different from a filter standpoint, and would #2 (cap to ground, resistor to VB) be a better arrangement in theory?

[ElectricDruid]

They're no different from a filter standpoint because what DC potential the bottom of the cap is connected to isn't really important since no DC can cross the cap.

However, from a noise point of view, the second one going to ground is preferred.

I agree with Jubal81 - it looks like a design that's been moved from a bipolar supply in the simplest way possible.

[Paul Marossy]

Paul said the caps were 47n and 220nF.    So as far as the circuit working it's like tube-screamer with the 4k7 resistor shorted.  So the gain at high frequencies is really high.

As you said, the DC connection of those resistors from the opamp -input will work.   In a tube-screamer type circuit the gain tapers to unity at DC.  At low *frequencies* the clipper is disabled.  With the DC connected Vref resistors in place the DC gain is no longer unity and the clipper is never really disabled, it's only disabled by a low input level.   I can only guess the motive is to reduce the amount of "clean" signal that leaks through the clipper.

I wonder if that is the principle behind the Peavey "saturation" control.

[Rob Strand]

I wonder if that is the principle behind the Peavey "saturation" control.

The Peavey ckt is more complex.  It has the saturation part in *cascade* with the gain part.
The main purpose of the gain part is to just change the gain (by forming a divider in the
feedback loop).  The fact it does provide a resistive path to ground is more incidental.
If they tried to decouple to two stages more with higher impedance (like when you cascade two filters)
it would end-up causing noise.   The saturation part is more or less independent of the gain part.
In the saturation mode the "TS-9 type" roll-off is always in circuit; however the amount of shelving is
affected by the Sat control.

[Paul Marossy]

OK, so here is the conclusion of these components in my (FINALLY!!!) working circuit:

1 - The RC filter thingies connected to the inverting inputs seem to have a dual purpose. They act as a filter, but they also seem to be involved in the biasing of the opamp somehow. If you take those out, the drive controls basically stop working. If you leave the cap out, then the drive pots do weird things and it sounds mis-biased and like c.r.a.p. If you have both the cap and the resistor connected (resistor to vref and then the cap), then everything works like you would expect it to.

2 - That resistor and cap in parallel connected to vref on the second stage non-inverting input, if you take that out, the circuit stops working. So that must be setting the bias on that input. That cap in parallel with it appears to filter out harsh harmonics from the first gain stage. I have that cap in my bread boarded circuit and that sounds good, but I left it out on my boxed up build. Without that cap in there, that thing is a little too bright... if I go past 30% of full counter clockwise it gets harsh and you can hear kind of a funny harmonic in the background - and that's going thru my kinda dark DIY Marshall speaker cab simulator. I also looked at the wave forms on the scope from my boxed up one and they are strange. When the treble is turned up there are some large needle-like spikes, never seen anything like them before. I will add that cap and then compare that to what they looked like before.

(BTW, if I take that cap to ground instead, it dulls the sound so I left it connected to vref like the original design)

3 - The comments about connecting to ground vs connecting to vref - the circuit as is is not noisy at all considering the amount of gain that it has.

4 - The comments about this looking like a dual supply design being converted to a single supply, conceivably that could be the case, but in this design I believe everything that is done was definitely intentional and not some kind of hack. I say that because of some of the people involved in the design of this thing, they're not some low level self-taught Joe Schmo in his garage hacking a circuit (like me ).

[Rob Strand]

If you leave the cap out, then the drive pots do weird things and it sounds mis-biased and like c.r.a.p. If you have both the cap and the resistor connected (resistor to vref and then the cap), then everything works like you would expect it to.

Really?  In your schematic is VR the same as Vref?   If they are different points/voltage that make sense.

What are the resistance values going to the opamp -ve inputs?

[Paul Marossy]

If you leave the cap out, then the drive pots do weird things and it sounds mis-biased and like c.r.a.p. If you have both the cap and the resistor connected (resistor to vref and then the cap), then everything works like you would expect it to.

Really?  In your schematic is VR the same as Vref?   If they are different points/voltage that make sense.

What are the resistance values going to the opamp -ve inputs?

Yeah, VR = Vref

One is 16K with .047uF (1st stage) and the other one is 14K with .047uF (2nd stage).

That's what is perplexing to me on this circuit and prompted my initial post - I was trying to understand what the hell was going on. You would think that it wouldn't matter if those RC filters are connected or not, it ought to still work, or at least that's how I see it. But it does matter in this circuit if they're not there. And you can't just stick vref right on the inverting inputs because it screws up the biasing. It's just a weird S.O.B. 

Don't worry, I'm sure the full schematic will soon be revealed. Then you can study it and understand it better than I do.

EDIT: Original schematic called for JRC4558s, I used JRC4580s. On bread board I am using combo of JRC4580 and 1458. Same results. I was thinking maybe specific opamp used might matter but it doesn't seem to care what specifically is used for IC-1.

[Rob Strand]

Yeah, VR = Vref

One is 16K with .047uF (1st stage) and the other one is 14K with .047uF (2nd stage).

That's what is perplexing to me on this circuit and prompted my initial post - I was trying to understand what the hell was going on. You would think that it wouldn't matter if those RC filters are connected or not, it ought to still work, or at least that's how I see it. But it does matter in this circuit if they're not there. And you can't just stick vref right on the inverting inputs because it screws up the biasing. It's just a weird S.O.B. 

Don't worry, I'm sure the full schematic will soon be revealed. Then you can study it and understand it better than I do.

Tell you the truth, your list of points are very puzzling to me too.

The resistor values don't look like they should cause any weird behaviour.    I was thinking the input bias current of the opamp was being use to deliberately create a DC offset.  I suppose measuring the DC voltages at the output of the opamps for different drive settings might point to something.   The 14k + 16k resistors do add some LF gain perhaps giving it a thicker tone.

There's a clue or detail missing.   When we know that it will all be obvious 

[Paul Marossy]

The resistor values don't look like they should cause any weird behaviour.    I was thinking the input bias current of the opamp was being use to deliberately create a DC offset.  I suppose measuring the DC voltages at the output of the opamps for different drive settings might point to something.   The 14k + 16k resistors do add some LF gain perhaps giving it a thicker tone.

There's a clue or detail missing.   When we know that it will all be obvious 

Well there is that whole CD4066 CMOS tone control thing we discussed last week which is part of this same circuit. Don't know if that factors into this or not.

[Rob Strand]

Well there is that whole CD4066 CMOS tone control thing we discussed last week which is part of this same circuit. Don't know if that factors into this or not.

At first through I wouldn't think so but you never know.   There's something wacky about those resistors.  The will probably explain why they are there in the first place.

I'm not quite switched on today so I probably shouldn't be posting stuff.

[Paul Marossy]

Here's a quick video of those wave forms. The spikes were bigger before I added the aforementioned 0.22uF cap

[duck_arse]

nice bird, thanks for showing him/her. [does he talk?]

[j_flanders]

The spikes ...
From the youtube video: "never seen anything like that.

Sorry to say this again, but I see this all the time. It's just a very common 'thing'. Maybe I'm not getting what you're hinting at.
In case I did, here's a random example. (I used this one because I knew I'd find this the fastest on the internet):

...more in the link below (or just about any square wave/ (hard) clipping followed by tone stack/control graph/plot on the internet)

By coincidence from someone at this forum...
http://tech.thetonegod.com/bmp/bmp.html

Here's another example, just in case the BMP is considered an exception.
Simulation of a solid state Marshall amp I did in Falstad.com:

[Paul Marossy]

nice bird, thanks for showing him/her. [does he talk?]

I think he tries to, but it's nothing like you would think of when you imagine a clearly talking parrot.

[Paul Marossy]

The spikes ...
From the youtube video: "never seen anything like that.

Sorry to say this again, but I see this all the time. It's just a very common 'thing'. Maybe I'm not getting what you're hinting at.
In case I did, here's a random example. (I used this one because I knew I'd find this the fastest on the internet):

...more in the link below (or just about any square wave/ (hard) clipping followed by tone stack/control graph/plot on the internet)

By coincidence from someone at this forum...
http://tech.thetonegod.com/bmp/bmp.html

Interesting. Well, first of all, I haven't been doing much DIY anything for about the last 4 years, so I'm not up on everything like I ought to be. Second, I personally have never seen this on a scope before, direct from the output of the pedal. I've looked at wave forms from a bunch of things I've built and have never seen this one before coming out of the output jack. Maybe I have somewhere in the middle of the circuit but never at the output jack. So I guess it would appear that this is "normal"?

What I am wondering specifically is what exactly are those spikes? Like massive 3rd order harmonics? Also at some settings I see something that looks like a mild crossover distortion, but it's near the top of the wave form.

[j_flanders]

So I guess it would appear that this is "normal"?

I'm just a noob in all this. But from the many circuits I've simmed and the few amps and pedals I've built or modded that is what I usually see when there's a lot of gain/clipping followed by a tone (RC) section.
I can't explain or interpret it, I'll leave that to the experts. As I said, I'm an absolute beginner in all this.
But if I were to be asked what a (heavily) clipped and tone shaped (towards the trebly side) sine wave would look like, that's what I'd draw. Which is why I immediately recognized it.
I added another example to my post above.

Lovely bird btw!

[Paul Marossy]

So I guess it would appear that this is "normal"?

I'm just a noob in all this. But from the many circuits I've simmed and the few amps and pedals I've built or modded that is what I usually see when there's a lot of gain/clipping followed by a tone (RC) section.
I can't explain or interpretet it, I'll leave that to the experts. As I said, I'm an absolute beginner in all this.
But if I were to be asked what a (heavily) clipped and tone shaped (towards the trebly side) sine wave would look like, that's what I'd draw. Which is why I immediately recognized it.
I added another example to my post above.

Lovely bird btw!

Ha ha, I normally rely on my ears/experience and don't do the math much or look at the scope very often. This just happens to be one of those times when the two did not match at all (ears matching up with what I see on a scope). Anyway, I have learned a few things from this project that I didn't know before.

The green cheek conure, that's my male, Ringo. The female, Ollie, is currently sitting on five eggs, which this time maybe will be a success. She's laid twice before and they were duds, so hopefully this time something happens.

[Rob Strand]

The spikes arise are because the Muff control is boosting the treble.  If you use a low of low-pass filters to clean-up overdrives then you might not see it often.

If you think of the output of the clipper as an approximate square wave, treble boosting a square-wave has peaks on the edges.

The suck-out after the edge is the mid-cut.

Here a cool pic on EQ'ing square waves,

[ElectricDruid]

The other thing to remember is that wherever filtering is happening, you've also got phase shifts. And those phase shifts will affect different frequencies in the signal differently. For a complex waveform like a distorted guitar, that can make it a lot different from what we might "expect" to see.

These waveform pictures are all "time domain" - what the signal does at each moment in time, but for filtering, we'd be as well to look at the "frequency domain" - the amount of each frequency in the signal across the audio range.

Because of the phase shifts, the relationship between the frequency domain and the time domain (the waveform image) is not at all obvious. You can make a thousand waveforms that sound *exactly* like a square wave, and yet don't look anything like one, just by changing the phases a bit.