Research Project: Need your op-amp reviews!

[zagisrule!]

Hi everyone!

I am doing a research paper about the influence of various op-amps in audio circuits ie: How do they sound?

Anyway, I will be blind testing the following op-amps to various people and seeing what they sound like to test subjects.

uA741
LM358
OPA134
OPA627
AD797

If any of you all have opinions on the above units, please reply and let me know how they sound.

I have done quite a bit of searching, but if there is a thread or website like this already, a link would be great!



-Matt

 

[jcx]

http://tangentsoft.net/audio/opamps.html

but I really don't believe that the subjective evaluations would stand up to contolled psychoacoustic testing

you should also consider reading up on A/B/X double blind test protocols and plan on matching listening levels to better than 0.1 dB between setups


http://www.diyaudio.com/forums/showthread.php?postid=152392#post152392

suggests that much larger technology differnces between power amps can't be distinguished when levels and output impedances are matched

 

[zagisrule!]

I have already constructed an A/B/X switchbox containing two identical stereo op-amp circuits.

I have read about double-blind testing, but am using the subjective reviews to construct my argument that audiophiles believe they can hear differences between the units.


Thanks for your input.





-Matt

 

[jcx]

double blind, controls and level matching are Necessary for meaningful Subjective evaluation - not exclusive of subjective evaluation

Book:

http://www.wiley.com/WileyCDA/WileyTitle/productCd-0470869232.html

uncontrolled, anecdotal commentary on the net may be a source of ideas for subjective categories/subjecive "axis" for test and candidate op amps but only a controlled study with good uniform protocol can yeild rankings/results that can reasonably be compared between tested chips

A big problem with uncontrolled subjective commentaty is that some audiophiles don't know about level matching and Loudness curves - they regularly make comments that could be explained by precived frequency response varions that are expected as a result of not matching listening levels

 

[SY]

Let's cut the scope of this down to a size where you can actually do a decent experiment. Pick two of those op-amps and ask the question, "Can they be distinguished?" There are two obvious test routes to take. First, a conventional ABX switchbox sort. These tests have shown great sensitivity to things like level and frequency response; some are skeptical that this method is less sensitive to other phenomena.

An interesting and less-used alternative is to recognize that a box of gain using one of those op-amps isn't very expensive or difficult to build. You might want to build 20 boxes, half with one op-amp, half with the other, and the boxes randomly coded. Now you can start to have some fun...

 

[Minion]

You should also consider that different opamps will perform differently Based on the Curcuit, Pluss you can"t really compare Fet Input opamp (EGpa134) with a Bipolar opamp (EG:LM358) in the Same curcut without possible some modifications....

Some opamps might need some external compensation to work in the same curcuit as other opamps...

So I would suggest that you build a curcuit customized for each opamp to work at it"s full potential if you want the Testing to be Fair in terms of what opamp sounds best....


Cheers