AW: Re: LF411 Op-Amp dilemma

brad sanders radioactive at geocities.com
Wed Jan 22 20:16:46 CET 1997


>Hey, this is confusing. Someone has to post a list of
>common opamps, with their main parameters
>like offset, slew rate, input impedance, drive capabillity!

Just pick up some data books. Virtually ALL the manufacturers have WEB
pages; go there, and request data books. Soon you'll have more data
than you know what to do with!

Analog Devices has a good online resource, and a good product
selector. Both have "flowcharts" guiding to opamp based on
applications and parameters. Also, ADs web page will suggest
replacements for other maker's opamps. For example, if you go there,
type "CA3140"  and select "Harris" it will tell you AD doesn't make an
*exact* replacement, but you might want to look at the LF441 (or
something like this; I think it's the LF441 - anyway, the LF411 is
still in ADs databook, and they certainly don't suggest you try to
replace the 3140 with one...)

>But the TL081 has rather poor offset values.
>So my suggestion would be: Use OP-07 where offset is important,
>but not slew rate: S&H stages.
>And use TL081 (or its low noise version TL071) where offset is no
>problem, but high slew rate is important: Audio stages.

Yes - but I'd still only suggest the TL081 in AUDIO stages where the
input impedance needs to be spectacularly high (like the ladder
buffers in a VCF) and/or the source impedance is very high (again,
like in a VCF - or when buffering tube stages, or sensing apps in
expo/linear I/V convertors).

The TL081 has very low noise current, which means the source impedance
will be less of a factor in total circuit noise (remember: current x
resistance = voltage - and input bias and noise currents are
multiplied by source impedance) but most all FET input amps have high
input noise voltage. In the TL081 this comes out to about 13nV/^Hz vs
about 3nV/^Hz in the 5534. 

In most apps, this is a good 10db difference in S/N ratio. Given the
even HIGHER noise floor of the LF411 (25nV/^Hz) it's not hard to
imagine nearly a 20db S/N improvement JUST from better opamps!

>PS.: I knew about the poor output stage of the CA3140, but I didn't
>       know of general poor adio quality. Can you tell me what you mean?

The 3140 has a hideous little output stage. This is fine if you WANT
the "sound" to vary with load (like the electronote app mentioned) -
but for most audio work it is bad, bad, bad. No output drive,
asymmetrical slew performance, latchup prone input stage, high
harmonic and IM distortion (look up Jung's report on this in "Audio"
circa mid-70s) - generally it's just "not good." Once upon a time the
3140 was useful in low input current/offset apps and where you needed
to conserve batteries - but now there are skillions of opamps that'll
even do this better.

Replacing an LF411 with a CA3140 is a little like replacing a Hudson
with an Edsel.


> http://radioactive.home.ml.org <-------------------------------



More information about the Synth-diy mailing list