A couple noob questions about these two non-inv. op amp topologies

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

AVA

Well-known member
Joined
Jun 7, 2021
Messages
47
Location
Canada
Hi all,

As usual, in the morning, instead of scrolling through Facebook or something like a normal person, I'll look up various simple audio circuits and try to follow each component in order, and pick out what it's doing and why the designer might have chosen that specific component.
Call it self educating...
or insomnia or whatever

In regards to non inverting op amps, it's usually pretty easy to do.

An example is the Tube Screamer's clipping circuit I attached below.

One question spread over 3 points:

- I often see small capacitors (C4 in the Tube Screamer) in the feedback of op amps, I understand this is to negate the buildup of parasitic capacitance, and that the value is generally small because it will create a low pass filter - only on the signal that is passed through it. (right?)
- Looking on mouser, the only logical type of capacitor to purchase for this is either a ceramic or MLCC because they seem to balance cost, size and longevity. However, putting a ceramic or mlcc in the audio path makes me uneasy, since I've seen all the heated online debates about capacitors... which I hope to never get into.
- Is this totally unfounded? Or is there an alternative solution that doesn't cost a fortune both in money and PCB realestate? How integral is this component to the audio signal really? Cuz it sure is tempting to just use an MLCC like I have been for the last while... But I have noticed some undesirable 'crackling' lately too. Hmm



I also stumbled upon the famous Klon Centaur circuit the other day and was quite intrigued by the design.

Something confuses me here, and it shouldn't...which means it really bothers me.

In the feedback of that Op amp you will notice R11||C7, and R12||C8 all making (?high pass?) filters, which is fine and dandy...

It was always my understanding that this network eventually needs to lead to ground, however you can see that it goes through the Gain pot, and back into it's non inv input, picking up Vcc/2 along it's way...which queues a few questions:

- How does this not change the DC biasing for the op amp? When the gain is down, the biasing is where?
- How can this actually create gain without just hitting the rails?
- There are two 'boxes' of filters; one is tied directly from the output to inverting input, and one is in between that and the non inv. input...How do these react differently with the gain? I'm assuming that the R11||C7 combo is always acting on the signal, while the R12||C8 combo is only acting on the 'boosted' portion of the signal...is that correct?

Please excuse my poor nomenclature on the topic, if I've made some mistakes in that regard.

It would be incredible if somebody could start at C5, or at the non inv input and just walk me through the klon circuit a little bit.

As always, I appreciate you and your time!!
I hope you're well


Alex
 

Attachments

  • tube-screamer.png
    tube-screamer.png
    22.2 KB · Views: 45
  • klon-centaur.png
    klon-centaur.png
    329.4 KB · Views: 45
The small value capacitors are small enough that you can easily use Class 1 C0G / NPO dielectric type capacitors.
Essentially, These don't have the same voltage and temperature coefficient issues of class 2 etc types like X7R , X5R, Y5V, Z5U.

For the Klon circuit - I'll suggest the best thing to do to get a better understanding would be to simulate it in LT Spice (as it's free).
But you may want to start with some simpler circuit blocks such as Sallen & Key and MFB active filters.
 
Last edited:
Yes, the small capacitor is meant to roll off gain at high frequencies and compensate for parasitic capacitance, like Newmarket mentioned, you can use C0G / NPO, which are amazing capacitors, I recommend reading Cyril Bateman's series of articles on capacitor distortion.
 
The small value capacitors are small enough that you can easily use Class 1 C0G / NPO dielectric type capacitors.
Essentially, These don't have the same voltage and temperature coefficient issues of class 2 etc types like X7R , X5R, Y5V, Z5U.

For the Klon circuit - I'll suggest the best thing to do to get a better understanding would be to simulate it in LT Spice (as it's free).
But you may want to start with some simpler circuit blocks such as Sallen & Key and MFB active filters.
Thanks for your time!

I love ltspice haha, but I'm on a Mac and it's super frustrating.
Recently I accidentally deleted a folder that contained all the opamps I curated and since then I've just been going straight for the breadboard...
Sounds like I'll have to spend some time putting ltspice back together again.

Cheers! Thanks for the tips
 
Yes, the small capacitor is meant to roll off gain at high frequencies and compensate for parasitic capacitance, like Newmarket mentioned, you can use C0G / NPO, which are amazing capacitors, I recommend reading Cyril Bateman's series of articles on capacitor distortion.
I will check this out! Thanks for the recommendation!
 
Thanks for your time!

I love ltspice haha, but I'm on a Mac and it's super frustrating.
Recently I accidentally deleted a folder that contained all the opamps I curated and since then I've just been going straight for the breadboard...
Sounds like I'll have to spend some time putting ltspice back together again.

Cheers! Thanks for the tips
I keep all my really useful stuff on a USB stick. You can get ones as big as 128GB quite cheaply now. I don't have to worry about computer crashes any more and I can take it with me. It even survived running through the washing machine and the tumble dryer :)

Cheers

Ian
 
Hi all,

As usual, in the morning, instead of scrolling through Facebook or something like a normal person, I'll look up various simple audio circuits and try to follow each component in order, and pick out what it's doing and why the designer might have chosen that specific component.
Call it self educating...
or insomnia or whatever

In regards to non inverting op amps, it's usually pretty easy to do.

An example is the Tube Screamer's clipping circuit I attached below.

One question spread over 3 points:

- I often see small capacitors (C4 in the Tube Screamer) in the feedback of op amps, I understand this is to negate the buildup of parasitic capacitance, and that the value is generally small because it will create a low pass filter - only on the signal that is passed through it. (right?)
yes, some lead capacitance across the feedback network will compensate for stray lag capacitance to ground. This use of feedback capacitance generally improves stability with the exception of the few op amps that are not unity gain stable.
- Looking on mouser, the only logical type of capacitor to purchase for this is either a ceramic or MLCC because they seem to balance cost, size and longevity. However, putting a ceramic or mlcc in the audio path makes me uneasy, since I've seen all the heated online debates about capacitors... which I hope to never get into.
- Is this totally unfounded? Or is there an alternative solution that doesn't cost a fortune both in money and PCB realestate? How integral is this component to the audio signal really? Cuz it sure is tempting to just use an MLCC like I have been for the last while... But I have noticed some undesirable 'crackling' lately too. Hmm
Some people argue that capacitor linearity doesn't matter for feedback poles tuned way above the audio passband but.... HF IMD (intermodulation distortion) can generate much lower and very audible distortion artifacts from HF signals. (this generally sounds like LF mud often occurring under loud HF signals like cymbal crashes).

As others have already shared cog/npo capacitor dielectric is very linear.
I also stumbled upon the famous Klon Centaur circuit the other day and was quite intrigued by the design.

Something confuses me here, and it shouldn't...which means it really bothers me.

In the feedback of that Op amp you will notice R11||C7, and R12||C8 all making (?high pass?) filters, which is fine and dandy...

It was always my understanding that this network eventually needs to lead to ground, however you can see that it goes through the Gain pot, and back into it's non inv input, picking up Vcc/2 along it's way...which queues a few questions:

- How does this not change the DC biasing for the op amp? When the gain is down, the biasing is where?
presumably the 4.5V voltage is a low impedance node so will not pass any Ac or DC past the wiper.
- How can this actually create gain without just hitting the rails?
by shunting the negative feedback network to an AC ground, increases the closed loop gain.
- There are two 'boxes' of filters; one is tied directly from the output to inverting input, and one is in between that and the non inv. input...How do these react differently with the gain? I'm assuming that the R11||C7 combo is always acting on the signal, while the R12||C8 combo is only acting on the 'boosted' portion of the signal...is that correct?
both legs of the NF network work together.
Please excuse my poor nomenclature on the topic, if I've made some mistakes in that regard.

It would be incredible if somebody could start at C5, or at the non inv input and just walk me through the klon circuit a little bit.

As always, I appreciate you and your time!!
I hope you're well


Alex
maybe start with easier circuits, and build upon that.

JR
 
I keep all my really useful stuff on a USB stick. You can get ones as big as 128GB quite cheaply now. I don't have to worry about computer crashes any more and I can take it with me. It even survived running through the washing machine and the tumble dryer :)

Cheers

Ian
This is a good call. I definitely won’t make that mistake again! Everything is backed up now so it should be wash machine proof.

Although I’m not sure my computer will survive the window I throw it out haha
 
Use the "Fahlstad" circuit simulator and copy the circuit in there and see what it does on the provided scopes.
Super Useful tool and regularly updated.
 

Latest posts

Back
Top