Are all op amps created equal?

[armdnrdy]

I would like to pull from this communities experience what differences have been encountered with different brands and "lots" of op amps.

Op amp manufacturers list the same numbers in their data sheets but have we encountered situations where certain products were found to be consistently lower or higher in some areas?

For example if the gain for a LM324 is listed across the board at min.50mv and typical 100mv, has anyone seen that ST Micro produced 324s are measured toward the lower end and Texas Instruments 324s measure toward the top end.

Some years ago I read a post on building a MXR flanger clone. It seems that there was an issue getting the build to flange with the "factory" CA1458 in place for the LFO. Once the op amp was changed to a 4558.....we got flange!

[R.G.]

Never ask for anecdotes, ask for data. They're different.

[armdnrdy]

Never ask for anecdotes, ask for data. They're different.

Okay...

Does anyone have any data suggesting that the same op amp produced by different manufacturers may vary in contrast to specifications listed on the data sheets?

[Kesh]

most data sheets have values saying min, typ and max. this covers all possibilities.

[armdnrdy]

most data sheets have values saying min, typ and max. this covers all possibilities.

Yes, I realize that. For a particular model, the data sheets seem to be carbon copies from one manufacturer to the next.
I was inquiring about real world data such as do Motorola LM358s generally have a noticeable higher gain than a Fairchild LM358.

[Mark Hammer]

The gain is not a function of the chip itself.  Do you mean a higher gain-bandwidth product?  a wider voltage swing?

[armdnrdy]

The gain is not a function of the chip itself.  Do you mean a higher gain-bandwidth product?  a wider voltage swing?

I guess depending on the application I could mean GBP or a wider voltage swing.

What I am striving for is any info as to a noticeable difference by switching op amps manufactured by different companies.
Does one manufacturer's op amps seem to "run hotter", produce more noticeable output in the same circuit?

Does the GBP vary greatly between manufacturers or different lots produced by the same company?

[Muthauzem]

One of the advantages of OpAmps designs over discreet components is exactly how you use huge amounts of feedback so you don't have to rely too much in their specs variation. As long as the gain is high and you application is in the audio frequency range, the results pretty much depend on the external components with most of the common OpAmps.

With that in mind, I would dare to say that if anyone make any kind of those claims here, he problably won't have enough data to support it. In other words, the only differences one could find would be "subjective" at best. Like "This one sound better than this one...".

We tend to believe the datasheet. I don't think any DIYer would have the time, money and reason to buy huge amounts of a component and test one-by-one if their specs meet with the datasheet.

[armdnrdy]

I was figuring with all of the tubescreamer builds to where every dual op amp under the sun was tried to find that "perfect" sound, that someone would have noticed if a particular one resulted in a "hotter" signal.

[Mark Hammer]

When you factor in all those tolerances of the diodes, the gain-setting resistors, the buffer transistors, the caps, etc., I don't know that anyone would be in any position to notice any difference in output level stemming purely from chip-choice.  There may be some tonal differences that appear to result in more presence, but that's not gain.

[armdnrdy]

Just wondering 

I'm looking at all angles for a possible fix for a build that doesn't sound quite right.

[Electron Tornado]

Regarding op amp comparisons in a guitar pedal, this is perhaps the closest thing to data that I've seen. http://www.youtube.com/watch?v=bpTv2jAree8

Anyone else have any data?

[gritz]

I'm afraid the only evidence I have is anecdotal. 

Years ago I was using an ST TL074 in a high gain distortion preamp design and I stumbled across a phenomenon called phase inversion - I was abusing the common mode input voltages and what I got sounded like a (broken) ring mod. Ugh! The TL 072's in my breadboard at the moment get abused horribly, but I've been unable to get them to pull the same stunt. So yeah, especially when you're pushing the boundaries it's possible that opamps from different manufacturers may behave differently. There's no reason for 'em to be internally identical and schematics in spec sheets are usually simplified as well.

Care to share the application you're debugging?

[R.G.]

The short answer is that yes, some very technical people have labored mightily for decades to make opamps incrementally better. The original opamps were vacuum tube devices. The first commercial IC ones were ... um? 709?... then the now-viewed-as-dismal 741. There are many, many subtle differences, all of which seem to be what happens out on the fringes of big signal, near power supplies, in and out of clipping, accuracy at DC, amplification at high frequencies, slew rates, gain-bandwidth product, bias currents, equivalent noise, and so on for many, many textbooks worth.

The problem is figuring out how that affects sound coming through them. It is not easy to find and collate data for this.

I recommend "Audio IC Op-Amp Applications" by Walter Jung. If you read this, you'll start out thinking you've found the Holy Grail, then about midway through say to yourself "... but how does that relate to pedal sounds??" and wind up at the end much more educated, but about the same amount frustrated.

I'm looking at all angles for a possible fix for a build that doesn't sound quite right.

Having been through a lot of this, I'd say to look at things other than the opamps for "doesn't sound quite right" first. It may be a quirk of the opamps, but I have only very, very rarely found that the opamps were the problem in a design. It may well have been the opamp doing exactly what I asked it to do - but that not being what I really wanted.