Op amp phase performance...

[ash_dac]

Hi,

MC33078:- The phase performance of this chip is terrible ?

http://www.onsemi.com/pub/Collateral/MC33078-D.PDF

I accidently had my speakers in reverse phase, and it sounded really strange. Am I right in this op amp will sound a bit odd due to phase change in audio band ?

Also the Lm833 has poor phase performance.

I have looked at the AD712, and this has a 'flat' phase performance in the audio band:-

http://www.analog.com/UploadedFiles/...399AD712_e.pdf

AD797, LM6172 seem to have flat phase.

Should I only select op amps with flat phase in the audio band ?

[Eva]

I'm sorry to say that but the only terrible thing here is your understanding of op-amps.

These graphs show open-loop phase response, and show it in different ways because AD712 datasheet shows phase margin (that's 180 degrees minus phase shift) while MC33078 datasheet shows bare phase shift.

Also, if you could actually understand a little these graphs and datasheets, you would notice that both MC33078 and AD712 have similar phase margins of approx. 50 degrees around their rated unity gain frequencies. However, these frequencies are 16Mhz for the MC33078 and only 4Mhz for the AD712, thus the former will always show less phase shift than the later at 20Khz in closed loop applications.

There is a lot of op-amp theory published in the web that you could use to gain better understanding about them.

[Bazukaz]

Phase shift itself is not a problem with music.
If you reversed phase on one speaker only , bass response was cancelling out.If you reverse on both , they should sound the same.
Phase shift also depends on circuit , not only the op- amp.
In active filters , it would look quite different from that you found in datasheet , and will not be linear anyway.

[ash_dac]

Originally posted by Eva
I'm sorry to say that but the only terrible thing here is your understanding of op-amps.

There is a lot of op-amp theory published in the web that you could use to gain better understanding about them.

Hi,

Thanks I will have a look.

In a lot of the threads the op amps seem to be parachuted in. I will have a dig around.


Many thanks,

Ashley.

[Eva]

Of course, there is a lot of people just swapping expensive op-amps (without even reading the datasheets) and trying to find "sonic excellence" that way, but that's quite far from science. No op-amp understanding is required to do that, a gentle budget and a lot of imagination are just enough.

[ash_dac]

Hi,

I've started to read this document:-

http://www.web-ee.com/primers/files/slod006b.pdf

[Eva]

It seems like a good reading, maybe a bit dense depending on your background. Chapter 7 may help you to understand what the phase plots that you saw in the datasheets are actually telling.

[poobah]

That is good start.

You must keep in mind, the manufacturer's are trying to describe the device itself and NOT the circuit you build around it. You notice the gain is probably 1,000,000 or more for these amps... where would you use that?

The circuit you put around the amp determines (mostly) its final gain & phase response. That is NOT to say the factors on the data sheet don't matter for audio... they do...

Keep doing your homework!

Good luck,

[Workhorse]

Originally posted by Eva
Of course, there is a lot of people just swapping expensive op-amps (without even reading the datasheets) and trying to find "sonic excellence" that way, but that's quite far from science. No op-amp understanding is required to do that, a gentle budget and a lot of imagination are just enough.

No EVA,
BUT This requires an IC socket and the correct way of inserting in it as well

[lumanauw]

Hi, EVA,

I don't understand, why a "genius" opamp designer said that opamps are not for audio? What is the problem? Too much gain can be burned by feedback, so what is the worst case?

http://www.diyaudio.com/forums/showt...695#post893695