What's the problem with opamps?

[Rev E]

I've read several threads with comments (usually negative) about opamps. I don't know too much about electronics ... I'm more of "user" than a techie. What's so bad about opamps? Do they change the sound, usually for the worse? Do passive circuits not have opamps? Do tell kids...

[crazydoc]

Hi Rev
I'm no techie myself, but I just semi-ignorantly mess with things electronic. As far as I know, there's nothing intrinsically bad about operational amplifiers - in fact, I love 'em.

Here's a neat little primer on how they work:
http://www.tonmeister.ca/main/textbo...ronics/12.html

They can be made as integrated circuits, or out of discrete components. There are thousands of different ones available, from those not suitable for audio use for various reasons, to those specifically designed for audio. They run in cost from a few cents, to the hundred dollar range (or maybe more - I don't know).
I suppose if you use the wrong one in a circuit it may not work optimally, or be noisy, or what-have-you. That's why people are always talking about "dropping" an upgrade into a circuit (many IC types have standard pinout configurations and can be pin for pin replacements for each other, especially easy if they are mounted in sockets).
Many even high end preamps and other equipment use op amp circuitry.

By the way, Rev, I was perusing some IC specs today and came across what is possibly a message to one of your kin. If you see him (or her), let him know. And say hi to the Rev's A, C, D, and any others out there.

[Industrial]

Ignorance and superstitious is part of the problem, applicability of op-amps to the intended uses discusses herein at these audio forums is the other part of the problem (the bigger one).

EQ Magazine just ran a little ditty about the "Pre-Amps in Paradise" conference.... thats a convention where some of the worlds best preamp designers and recording engineers meet to discuss the design of preamps and to talk a little bit about the music and recording business.... anyway.... building preamps always has been part science and part art.... or at least there exists a large element in this world who seem to think that electronics is an art form (although as an electrical engineer, I kind of laugh at these people... because its usually the people with non-technical backgrounds [record produces, musicians, some recording engineers] who seem to regard the laws of physics as an "art form" (a pure absurdity).

You find all kinds of superstitious talk about tubes versus transistors, transformerless designs versus transformer designs, as well as talk about whether integrated circuits should be used in pre-amps...

IC's, are not "bad". It's just that the particular specifications for integrated circuits make them not always useful for designing amplifiers which need to handle high powered signals.

The fact of the matter is that IC's can handle TREMENDOUS amounts of gain (110 - 120dB) which is actually more gain than any of the best pre-amps of the market which usually only offer 20 - 65 dB of gain..... Rather, the problem with IC's is not their gain capabilities.... but they just can't handle the input power.

Discrete circuity certainly, by no means offers less distortion or noisee than IC's..... Quite the contrary!!!!!

Some of the best IC preamplifiers will produce noise floors FAR FAR less than the worlds best audio preamps. Their noise floor at > 80 - 100dB of gain may [much] less than a discrete component preamp operating at only 30 - 70 dB of gain.

Some of the lowest distortion and lowest noise circuits in the entire world are in "chip" form, and are used in sophisticated electronics (military, communications, scientific instrumentation, and otherwise) which require ULTRA-low noise floors or ultra-low distortion levels for sensing applications. We're talking noise floor levels which absolutely, and totally blow away the $3,500 'super-duper' preamps that you see from manufactureres like Millenium or Avalon or Aphex or Robert Massenburg Labs, etc...

[regebro]

There is nothing wrong with opamps. Purists don't like them for a variety of semi-religious reasons. One real argument is that they are typically class B amplifiers, which is not the purist choice. Bit it can be noted that many of the famous old classic preamps, including many Neve designs are also class B or AB, so that too seems mainly a question of religion.

There is only one rule: If it sounds good, it is good.

When it comes to opamps, the TI Burr-Brown line is generally hailed as sounding ecxellent.

[bgavin]

The INA163 IC appears tailor-made to match 200-ohm microphones. The circuit claims very low distortion and quiet performance.

TI provides a base schematic for building a mic preamp arouind the INA163. Rane publishes their schematic for the MS1b which is also built on the INA163, as are their MLM series of which I own two.

It appears the extra bells and whistles are external to the IC proper. A tasty 2dB gain switch such as found in the Grace 201, is certainly far more expensive than the IC chip. Add in the very best caps, etc, and this IC does the job. The Rane offering is price concious, and uses lesser quality/priced components.

[Rev E]

Thanks so far everyone. Very insightful stuff so far. I didn't know that we had such experts hanging around here.

[crazydoc]

Although this IC is not an op amp per se, it likely uses them in its circuitry. Looks like it would make an interesting pre.
http://focus.ti.com/docs/prod/folder...t/pga2500.html

[bgavin]

I'd like to know more about the current feedback devices, such as Grace Designs, and (I think) the Ted Fletcher preamps. Both tout their designs as being superior to voltage controlled preamps.

[Flatpicker]

Originally posted by Industrial
...IC's, are not "bad". It's just that the particular specifications for integrated circuits make them not always useful for designing amplifiers which need to handle high powered signals...

Great point!

...Rather, the problem with IC's is not their gain capabilities.... but they just can't handle the input power...

Uh, don't you mean "output" power?

Anyway that's the problem with ICs. To date, no company has manufactured one that does all the things a mic pre needs to do like have low noise, low distortion- especially at the critical harmonics, and enough output drive, as good as a discrete opamp can. Not to mention have "that" sound.

When they do, they'll put John Hardy, Fred Forssell, API and a few other guys out of business in that department.

[Flatpicker]

Originally posted by bgavin
I'd like to know more about the current feedback devices, such as Grace Designs, and (I think) the Ted Fletcher preamps. Both tout their designs as being superior to voltage controlled preamps.

Well, first you got to learn to recognize marketing hype when you see it!

Simply put, you basically have bipolar transistor amps which are current controlled or FET - “Field Effect Transistors”, which are voltage controlled. Therefore, the feed back in a bipolar amp would have to, at some point, be converted to current, and the feed back in a FET amp would have to be converted to voltage. I can give you excellent examples of both and even a hybrid of the two:

The Jensen 990 discrete opamp that is manufactured by John Hardy and used in his preamps is strictly a bipolar transistor (current) design. The 992 discrete opamp designed and manufactured by Fred Forssell and used in his preamps/mixers is strictly a FET transistor (voltage) design. The Great River MP-2/4 uses an opamp designed by Dan Kennedy that utilizes a FET transistor input stage, followed by a bipolar transistor output stage, thus composing a hybrid amp.

Which is better? Well, it would probably only depend on your application as they are all top notch sounding pres. My point to all this is that it's ridiculous to make a statement that one design is superior to the other, especially concerning mic pres where beauty is clearly in the ear of the beholder.

[bgavin]

Thanks for the update. I'm not an EE... my education is computer science.

The Grace design document makes for convincing reading, and did not strike me as marketing fluff. The underlying theme is for accuracy, and my research indictes the Grace units have achieved that. Other terms I've seen are clinical and sterile. One man's clean is another man's sterile, I suppose.

Grace makes the point about voltage controlled circuits being less accurate. Perhaps this is where the coloration comes in? Speculation on my part.

[Flatpicker]

Originally posted by bgavin
Thanks for the update. I'm not an EE... my education is computer science.

The Grace design document makes for convincing reading, and did not strike me as marketing fluff. The underlying theme is for accuracy, and my research indictes the Grace units have achieved that. Other terms I've seen are clinical and sterile. One man's clean is another man's sterile, I suppose.

Grace makes the point about voltage controlled circuits being less accurate. Perhaps this is where the coloration comes in? Speculation on my part.

Ah, the old preverbal straight wire with gain. Thing is, nobody can say for sure what that really sounds like, simply because there ain’t one.

According to all who've used it, the Forssell 992 is a very accurate opamp. Its actually an improved version of the Millennia Media MM99, which Fred also designed. The M99 is also known to be of the "uncolored" and "accurate" variety. This is one reason I take the stuff from Grace with a grain of salt. But then again, whose to say they can't design a sterile sounding pre and call it accurate? A competitor could advertise “Accurate, but not sterile like current feed-back designs” and then explain how accurate voltage feed-back is. You see, one could build an argument both ways.

All hype aside, it just boils down to what you want to track with.

[acorec]

Originally posted by Industrial
Ignorance and superstitious is part of the problem, applicability of op-amps to the intended uses discusses herein at these audio forums is the other part of the problem (the bigger one).

EQ Magazine just ran a little ditty about the "Pre-Amps in Paradise" conference.... thats a convention where some of the worlds best preamp designers and recording engineers meet to discuss the design of preamps and to talk a little bit about the music and recording business.... anyway.... building preamps always has been part science and part art.... or at least there exists a large element in this world who seem to think that electronics is an art form (although as an electrical engineer, I kind of laugh at these people... because its usually the people with non-technical backgrounds [record produces, musicians, some recording engineers] who seem to regard the laws of physics as an "art form" (a pure absurdity).

You find all kinds of superstitious talk about tubes versus transistors, transformerless designs versus transformer designs, as well as talk about whether integrated circuits should be used in pre-amps...

IC's, are not "bad". It's just that the particular specifications for integrated circuits make them not always useful for designing amplifiers which need to handle high powered signals.

The fact of the matter is that IC's can handle TREMENDOUS amounts of gain (110 - 120dB) which is actually more gain than any of the best pre-amps of the market which usually only offer 20 - 65 dB of gain..... Rather, the problem with IC's is not their gain capabilities.... but they just can't handle the input power.

Discrete circuity certainly, by no means offers less distortion or noisee than IC's..... Quite the contrary!!!!!

Some of the best IC preamplifiers will produce noise floors FAR FAR less than the worlds best audio preamps. Their noise floor at > 80 - 100dB of gain may [much] less than a discrete component preamp operating at only 30 - 70 dB of gain.

Some of the lowest distortion and lowest noise circuits in the entire world are in "chip" form, and are used in sophisticated electronics (military, communications, scientific instrumentation, and otherwise) which require ULTRA-low noise floors or ultra-low distortion levels for sensing applications. We're talking noise floor levels which absolutely, and totally blow away the $3,500 'super-duper' preamps that you see from manufactureres like Millenium or Avalon or Aphex or Robert Massenburg Labs, etc...

God damn! The voice of reason. I have been saying this all along. The problem is with the people who don't understand what op-amps are and are capable of. Just because the worst performing pre-amps use them does NOT mean that a world class pre-amp cannot be designed with them.

[ausrock]

And I would strongly question anything coming from the (UK) Fletcher camp.