inverting vs noninverting opamp stages?

[caress]

just wanted to discuss the pros, cons and differences between using an inverting vs noninverting opamp stage.

let's start with some simple blocks that are seen everywhere:
*a buffer
*a gain stage
*a combination of the two - buffer > gain stage / gain stage > buffer

calling rg here... 

[CynicalMan]

1. A buffer.
A non-inverting circuit takes maybe 7 or 8 parts plus the op amp. An inverting buffer takes one more. Additionally, the inverting circuit will have either a low input impedance or a large resistor in series with the signal. Non-inverting wins here.

2. A gain stage.
Same issues as above, except the non-inverting one will often need one or two more parts than the inverting one. Still, it's worth it to solve the problems IMHO.

3. A combination of the above.
See the issues above for the first stage. The input impedance of the second stage is less critical, but there's another consideration here. If you run a buffer into a gain stage, any noise picked up in the first stage and between the two stages will be multiplied by the gain of the gain stage. If you run a gain stage into a buffer, you won't have that problem. So, I'd probably go non-inverting gain stage (for high input impedance) into a non-inverting buffer (for less parts).

I recall there being other differences but I can't remember what they are. I'm sure RG can help you with those.

[Johan]

inverting is usefull in high gain stages where oscillation due to positive feedback might be a problem, such as in distortion boxes..and allthough many people obsess about the signal being inverted and how it affects the sound, I have yet to hear about anyone actually having any problem due to it...besides, with the exeption of a few examples made for the american market that had a "phase"switch, all "Plexi" Marshall's are inverting, yet no one complains..
J

[cpm]

1. A buffer.
A non-inverting circuit takes maybe 7 or 8 parts plus the op amp. An inverting buffer takes one more. Additionally, the inverting circuit will have either a low input impedance or a large resistor in series with the signal. Non-inverting wins here.

2. A gain stage.
Same issues as above, except the non-inverting one will often need one or two more parts than the inverting one. Still, it's worth it to solve the problems IMHO.

3. A combination of the above.
See the issues above for the first stage. The input impedance of the second stage is less critical, but there's another consideration here. If you run a buffer into a gain stage, any noise picked up in the first stage and between the two stages will be multiplied by the gain of the gain stage. If you run a gain stage into a buffer, you won't have that problem. So, I'd probably go non-inverting gain stage (for high input impedance) into a non-inverting buffer (for less parts).

I recall there being other differences but I can't remember what they are. I'm sure RG can help you with those.

If you are doing a high gainer, a very hig input impedance will help in picking external noise (hum, radio, etc), wouldnt it? I'd think that a moderately low impedance input can be a good trick in that situation. If there's some highs lossing due to the low impedance input, i wouldnt care too much, i'd have some low pass in the gain stages anyway

[CynicalMan]

Maybe so, but that's easy to change without adding parts. Reduce the resistor to Vr and you're done!

[ashcat_lt]

I like this question and want to hear more responses.

I can offer one thing, though it's not really related to the specific questions above: 
An inverting stage is capable of gain less than unity.

[cpm]

I like this question and want to hear more responses.

I can offer one thing, though it's not really related to the specific questions above: 
An inverting stage is capable of gain less than unity.

and thus can be used as a simple DC level shifter, up and down

[jaysg]

You need to invert if you're going to create a mixer.   If I had time for it, I would play around with spliting the signal into frequency ranges like an EQ does, selectively distorting them, and then mixing them back.

[askwho69]

i like the topic

[merlinb]

just wanted to discuss the pros, cons and differences between using an inverting vs noninverting opamp stage.

A buffer and a non-inverting gain stage are basically the same. One just provides more gain, so apart from that there's not much to choose between the two.

An inverting gain stage is useful as a volume control since it's gain can be reduced all the way to zero, and it is also useful as a mixer.

However, an inverting stage has the problem of low input impedance. And if you try to get a high input impedance by using large resistances then you get lots of Johnson noise (hiss!)

To get around this you need to add a buffer before the inverting stage. This way you get the high input impedance of the buffer, and you are free to use small resistances in the following inverting gain stage (since the buffer can drive heavier loads than your guitar can), meaning a lot less overall noise (even though you're using more opamps!)

[ayayay!]

Merlinb, that was so good I copied it and made it a document to keep handy. 

[Derringer]

man .... i can't remember the link at all
but I'm pretty sure there's a comparison somewhere between and inverting vs non-inverting opamp circuit with diodes in the feedback loop
and there is a difference between the output signals ... in how they clip

[waltk]

This is a wonderful app note document from TI: http://focus.ti.com/lit/an/sloa058/sloa058.pdf

It's titled: "A Single-Supply Op-Amp Circuit Collection", and is full of simple examples of inverting/non-inverting opamp circuits.

[CynicalMan]

man .... i can't remember the link at all
but I'm pretty sure there's a comparison somewhere between and inverting vs non-inverting opamp circuit with diodes in the feedback loop
and there is a difference between the output signals ... in how they clip

http://www.bteaudio.com/articles/TSS/TSS.html

[boogietone]

Been looking for a good summary of this. The tube screamer discussion is pretty fascinating.

[Derringer]

man .... i can't remember the link at all
but I'm pretty sure there's a comparison somewhere between and inverting vs non-inverting opamp circuit with diodes in the feedback loop
and there is a difference between the output signals ... in how they clip

http://www.bteaudio.com/articles/TSS/TSS.html

THANKYOU !

going to save the link this time