Opamp change Now some Disappointments

[Mooly]

It's not a figure that's normally quoted... high is the answer.

This might help explain it better,
http://www.analog.com/static/imported-files/tutorials/MT-040.pdf

 

[ChPuls]

It wouldn't make any difference in the circuit if the input resistance would change from 300k to 30k Ohm?
Ok both values are high.
What is about noise? There is current and voltage noise.

 

[Mooly]

Noise depends as much as anything on the actual circuit impedances used.

It's too huge a subject to sum up in a few words, but generally FET opamps are noiser than bjt's (voltage noise), but have less current noise... which you use depends on the exact design of the circuit as a whole.

 

[mchabbi]

And can someone recommend me an opamp that has good soundstage?
I already tried the LME49720 but the stage is only distant with no good different levels of depth.
I haven't broken in the LME49720. Does it improve over time?

Has someone compared different NJM/JRC like JRC4565 , JRC4580 or JRC2068?

JRC 4565 sounds much much better than 2068, slightly better than 4580.

 

[SoNic_real_one]

And all three of them worse than LM4562 (LM49720).

 

[linuxguru]

It is the small Kenwood DP-1001. It has the Philips TDA1547 (or DAC7) as the DAC. It is from 1994 and was then a good CD-Player. The opamps have even as default a good power supply! And they are followed by an unloading discret current stage.

I just upgraded a Kenwood DP-1001 with LF03d discrete opamps, and the difference is *huge*. The main issue is that the stage immediately following the TDA1547 DAC has relatively low rails of ~ +/- 5v or lower, so most highly regarded monolithics choke there.

http://www.diyaudio.com/forums/analog-line-level/207871-lf03-discrete-opamp-3.html#post3660276

I designed the LF03d for rail voltages ranging for +/-3V upwards, so it has no problems in the Kenwood DP-1001 (or other low rail applications like soundcards, etc.). With the LF03d, the DP-1001 utterly dominates the Marantz CD-6003, whereas it was not that obvious with high-performance monolithics like the LME49722 and 2x OPA627 in the I/V.

 

[auwdio]

I wonder how many of you are testing with fake opamps ordered at aliexpress, in my experiance almost all of them were relabeled opamps. probably lm358 orso.

 

[Extreme_Boky]

I only want to try a better opamp from NJM/JRC.

Maybe LM6172?

 

[diyralf]

You can't exchange an opamp for another type. This is the dumbest thing you can do. Do not do it again! A circuit that works cannot work with another opamp.

 

[jan.didden]

You can't exchange an opamp for another type. This is the dumbest thing you can do. Do not do it again! A circuit that works cannot work with another opamp.

Funny that you say that. I routinely swap opamps and almost all of them work. It is in the nature of opamps that their performance is largely determined by the external loops so the differences between types in an application are small.
But there are differences and sometimes it is of interest to find out exactly what these differences are, and to select the best one for your purpose.

Edit: what often happens is that people replace a 'normal' opamp by a high-speed one in the expectation it will sound better. Often this leads to oscillations because the circuit is not correctly compensated for high speed opamps. This makes it indeed sound different and the expectation bias makes it perceived as 'sounding better'.

Jan

 

[diyralf]

...and you have 250 times more THD after the replacement. All possible.

 

[phase]

I even use ones purchased off the auction sites from overseas and they still work well, somehow.