[sdiy] The sound of various Op-Amps

[Scott Bernardi]

That is another one of the trade-offs I mentioned earlier.  Low noise 
opamps, with an input stage running at a high current (like the 5532) 
are actually optimized for low voltage noise; current noise is higher.
Input bias current is probably the second main reason (other than slew) 
that the TL* series opamps are used a lot in synth building.  Used as an 
integrator in an oscillator circuit, for instance, the input bias 
current will determine the lowest frequency you can go to.


Scott Bernardi wrote:

    It turns out you also get lower noise running at a higher current,
    so you get the low noise and higher slew, both good for audio. This
    is at the expense of other parameters like input bias current (for
    bipolar inputs) or offset voltage and drift (for JFET or MOS
    inputs).  It's all a bunch of trade offs.


Roman wrote:

>bias current is one thing - on 10k input resistance
>500nA gives 5mV, although it can be trimmed.
>But high bias current opams usually have high input equivalent
>current noise. So if 5532 has at 30Hz 8nV/sqrt(Hz) voltage
>noise density, and 2.7pA/sqrt(Hz) current noise, one can calculate that when input resistance rises above 8nV/2.7pA=3k, current noise is contributing more noise to the circuit than voltage noise.
>So using input's higher than 3k means that its performance is not fully used, and at some point, probably with not much more resistance, it may become more noisy than 741.
>For super lowish noise opamps, thermal noise of the resistors is also important, and some applications require input impedance in range of several hundred ohms to keep the noise down.
>
>Roman
>
>---- Wiadomość Oryginalna ----
>Od: Andre Majorel <amajorel at teaser.fr>
>Do: synth-diy at dropmix.xs4all.nl
>Data: Mon, 22 Sep 2003 10:28:07 +0200
>Temat: Re: [sdiy] The sound of various Op-Amps
>
>  
>
><br />
>  
>
>>Thanks Harry, that's interesting. I understand the DC offset stuff
>>    
>>
><br />
>  
>
>>but what makes you say it's only good for low input impedance ? Is
>>    
>>
><br />
>  
>
>>it the high biasing current ? (500 nA for the 5532 vs. 65 pA for
>>    
>>
><br />
>  
>
>>the TL072)
>>    
>>
><br />
>  
>
>
>  
>

-- 
Scott Bernardi
sbernardi at comcast.net

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://synth-diy.org/pipermail/synth-diy/attachments/20030922/330102ca/attachment.htm>