Opamps for audio: What's the voltage noise king these days (late 2024)?

[cvanc]

Has anyone gone past the LT1115?  AFAIK that's still the quietest when discussing voltage noise?  Thanks.

www.analog.com/media/en/technical-documentation/data-sheets/lt1115fa.pdf

[Someone]

If only there was a carefully curated list of parts with their relevant parameters that could be searched and sorted to find out such information.....

[David Hess]

The comparable Texas Instruments part is the OPA1611/OPA1612.

For not quite as low noise but more useful higher source resistance, a JFET part like the OPA1641 or OPA1655 is better, or the CMOS OPA1677.

If you really need lower input voltage noise at any cost, then a discrete input stage or multiple operational amplifiers in parallel is still required.

[magic]

ADA4898 is supposed to come very close at a much higher speed.

[David Hess]

The ADA4898 achieves lower noise by not having input bias current cancellation, which should not be required in an audio application anyway, but is also a 2-stage design so I would expect higher distortion which seems to be born out in the specifications.  I guess they try to make up for that with a more linear input stage.

[magic]

For the record, everybody except Douglas Self calls this a single stage design.

Distortion is only speced above 100kHz, but the slopes are increasing so it should be better below 100kHz.

This seems related to AD797, which has the same voltage noise except near 20Hz, and has been tested by Samuel Groner to be one of the lowest distortion opamps out there. In particular, it is essentially unaffected by 600Ω loading and apparently drives 200Ω better than TI's OPA211 (industrial version of OPA1611) drives 600Ω.

[David Hess]

Distortion is only speced above 100kHz, but the slopes are increasing so it should be better below 100kHz.

Do not underestimate marketing.  If the slope changed below 100 kHz and distortion increased, they could leave it out of the datasheet.  The only way to really know is to test it.

[0xFFF0]

[MasterT]

I come accross this thread:
https://e2e.ti.com/support/amplifiers-group/amplifiers/f/amplifiers-forum/421088/op-amp-dirty-little-noise-secrets
 And indeed LT1028 chart shows :

 Same time LT1115 DS hide this misbehavior , but comparing Gain-Phase charts of both LT products:
it's obvious that LT1115 'd have same abnormal noise hump at 400 kHz.
 

Edit: picture 1 & 2 swapped, can't get it right.