[sdiy] Equiv of LM3900?

[Magnus Danielson]

From: harry <harrybissell at prodigy.net>
Subject: Re: [sdiy] Equiv of LM3900?
Date: Sat, 02 Mar 2002 21:13:48 -0500

> The big point here is "what did you want the opamp to do"...
> 
> The LM3900 has a very (like extremely) low input impedance.
> 
> Usually we select voltage input opamps to have a high impedance
> input...
> 
> so for some circuits you 'might' sub another opamp... but if the designer
> 
> really understood the 3900... you probably can'r because the reasons for
> that choice are almost incompatible with 'normal'  (not Norton ;^)
> opamps.

This is indeed true, there are certain designs that *really* require
their specified op-amp. For instance, Jürgen reported that he found
and amplifier design where the uA741 was required, due to the way
things where setup and the input transistor-pair was effectively
disabled and replaced by a external and better pair.

This is a very good reason why I would like to see good datasheets
(and similar info which gives insight into the device and its
operations) even after end-of-life for a component. You can't
fully understand what an engineer thought of unless you have the same
information as what was around at the time of the design (preferably
more info naturally).

The more one gets into the habbit of crawling into a chip, the better
one can possibly be at making it do The Right Thing (TM) but it also
opens up the path for Oversmart Hacks which usually is not The Right
Thing (TM).

Cheers,
Magnus