monolythic ic listening tests

[buttachunk]

i've spent the last 2 days making listening tests to monolythic ic's. for my test setup, i used a pair of MTM's (thor design by d'appolito) with morel tweeters and custom aluminum drivers i finished last week which sound INCREDIBLE.

i used a very transparent amp i know very well and rebuilt about 2 years back. all moster cable.

so my test bed is an older well-built cd player with a burr-brown DAC. i took the soldered-in 5532 output buffer out and replaced it with a socket. so this one single (and very important to the signal chain) opamp socket was the epicenter of the listening test.

all i can say below is subjective (EDIT;) based on this system in this circuit;

1. i'll never bash a 5532 again. they spec out over 3 times better than an API 312 and they honestly have much more presence and body than i realized... i was a 5532 basher, never will be again since it was in the top 3 sonically. the thing goes to 10MHZ-- i don't know why so many people say they think it sounds 'dark'.

2. burr brown 2604-- silky, smooth. not the choice for agressive material- definately THE monolythic choice for a tube-like sweetness. incredibly listenable.

3. national semi 6172-- firstly, this chip required an additional .01 poly to prevent oscillation (will find a better permanent solution in the future). this one sounded great as well-- very smooth and articulate, no harshness. actually sounded more open than the 5532 and had all of the presence and dynamics.


then, honestly, there was a startlingly huge gap. none of the other chips i listened to sounded even suitable in this circuit-- some even sounded subtly "eq'd". now i don't want to list model names because i don't want to deflate anyone.

to be fair, this output stage was designed for a 5532, so i'm sure the other chips would sound fine or better in other designs.

also note i definately didn't listen to every chip, so don't throw any chips away because of this DAC test ;) .

i recommend to everyone who questions these things to find a great monitoring system that they know well, and find a well-designed piece of gear which has one crucial opamp stage and just listen to several of the chips on the market and find the sonic additions of each.


[img]graemlins/beerchug.gif[/img]

[ February 11, 2004, 07:02 PM: Message edited by: buttachunk ]

[thermionic]

Hey Butta, have you seen this thread?

http://www.recording.org/ubb/ultimat...=001646#000000

[buttachunk]

justin

thanks for finding that one ! when i typed 5532 into the search, i think every thread came up. :D

[buttachunk]

has anyone compared different brands of 5532s ?

anyone else notice differences ?

tekay measured huge differences between brands making the 3055s...

i heard a difference in the jrc 5532 and the philips i used in the listening tests... the jrc sounded waxier and the philips sounded cleaner on this system.

[Bear's Gone Fission]

There's a thread on discrete vs. monolithic at the Geekslutz forum that touches on some of this. All the things that make these ic's different mean that, while pin compatible and while they do generally work when subbing eachother, that each will generally thrive best in circuits designed well to the specific chip. The trick is how many circuits we try to hop up that are poorly designed and/or are fitted with a lesser chip than the designer specified, because it works and it's cheaper.

I think sometimes we're just picking which opamp has the most pleasant distortion for a given substitution. (As I'm ordering sockets to redo my board . . .)

Bear

[tkohalmi]

Is it possible to see a schematic for the circuit that you used to evaluate them? In my experience the design and the purpose for using opamps vary tremendously. There may be more to learn from your experiment had we understood its details a bit more.

tamas
Fan of Opamps

[buttachunk]

tamas--

wish there was a schemo. it's a multilayer board... couldn't trace it with the test equip. i have here.

seems like an average output buffer... probably textbook DAC out.

it seems that the touchiest of the bunch electronically was the LM6172, and it sounded great with the polyprop cap. tried the same cap with other opamps, and the difference was kindof severe-- of the fast video amps, the LM6172 not only is the fastest, but is the best sounding of the bunch i listened to.

again, didn't listen to nearly every chip... and these speakers have already brought out alot of stuff i never heard before...

[spaceludwig]

Butta,

Output buffer - does it amplify? What is the gain?

[buttachunk]

space--

i think it's unity gain. the opamp is there to drive certain loads and to kindof protect the DAC.

i'm sure this low gain circuit is what really bent some of the chips... ;)

[ggoerss]

I`m constantly bemused by Audiophools who swap out a 5532 out of their cd players and replace it with something more esoteric.I wonder if they consider that in the process of recording, the signal has probably passed through 200 5532s, so as if 1 more is going to make any difference.
Your results tend to mirror the results posted by D Self on his website regarding tests between different opamps.The 5534/32 seems to cop alot of criticism most of it unfairly.They are still as relevant today in their usefullness as they were in the mid 70`s