[sdiy] Op amp notes

[Paul Higgins]

On Saturday, September 4, 2004, at 05:32 PM, Cary Roberts wrote:

>> Aren't nearly all boards, regardless of price, chock full of 
>> monolithics?
>
> Most new stuff, yes.  API and Daking are exceptions to the rule.

I bet you'll eventually find monolithics in even the APIs, won't you?  
For example, parametric EQ stages, etc?  I know for sure that lots of 
well-regarded British desks have tons of monolithics in them--just not 
in the mic preamp front-end.  (Well, I've read that the Neves are 
totally discrete, but how many others really are?  People go nuts over 
the Focusrite desks, and those can't possibly be all discrete, can 
they?).

> The stock  Sonys had the option of a discrete transformerless input 
> and a 5534
> based Jensen transformer based mic pre.

A 5534 following a Jensen transformer?!  As Roddy Piper's character in 
"They Live" said, "that's like pouring perfume on a pig!".  I've used 
the mu-metal double-shielded mic transformers from Sescom and 
Reichenbach Engineering (similar specs to Jensen) and been very happy, 
but I wouldn't think of using something other than a tube, exotic FET 
cascode circuit, discrete opamp, etc.  Well, to each his own...  (And 
before I get flamed for dissing the 5534, let me say that I use them 
all the time.  Just NOT for mic preamps).

> People still pay top dollar for discrete power sucking Neve, API, and 
> Quad Eight
> gear.  Just like folks still prefere Moog, EMS, EML, and ARP designs 
> to early 80s
> integrated designs.

Well, people prefer them for tracking.  You'd better have one hell of 
an automation system if you want to mix on them though!

> I'm not averse to using opamps in synth circuits - I don't think it's
> warranted to use a fancy discrete amp as current to voltage converters,
> buffers, etc.

It's hard to screw up a signal when it's 10Vp-p.  MHO is that it's hard 
to screw it up even at +4 dBm.  Of course, if you're trying to drive a 
600-ohm line with a 5532...

> The problem is it's not just ICs but also older transitors and such as 
> well.
> There are "equivalents" but they never sound the same as the original. 
>  Of
> course, with thirty year old parts, how much of that difference is due 
> to
> age vs manufacturing differences?

Wow...didn't know that about the discretes.  It might be similar to old 
tube amps, where part of the sound of those amps is that they have a 
zillion miles on them.  Naturally, no one ever listens when you try to 
tell them that you have to play the amp that long to get it to sound 
like that!

> I've become a switched filter circuit EQ snob.  I've been able to buy 
> 100+
> concentric Grayhill rotary switches on ebay for less than $1 each, and
> bought a box of Grayhill parts (enough to make 300+ switches) to 
> expand them
> to three or four pole.  It doesn't get nicer than just using resistors 
> and
> caps (and inductors depending on design) to pick EQ points and using a 
> few
> buffer amps (either IC or discrete) to finish out the design.  Gyrators
> don't bother me so much as many IC based EQs are constant Q not 
> constant
> power.

Man, that's hardcore.  I don't envy you in your attempts to get some of 
those old inductors.  Or do you wind your own?

-PRH