Wow: quite a difference from replacing the EQ section op-amp…

Hi all.

Recently, I've gone on an op-amp upgrading rampage. I basically
got rid of all audio-handling dual op-amp IC's in my ADA MP-1,
and replaced them with NE5532 chips from On Semiconductor.

What a fantastic difference in sound quality, and the unit is
incredibly quiet now. I can put the guitar down with the volume
up even on a high gain setting and the noise is negligible. From
this I learned that the MP-1's reputation for being a noisy
preamp is in fact due to numerous inferior op-amp chips.

Yesterday, I replaced one more op-amp, for which I had to order
the replacement. I kicked out the LF347 in the EQ section and
replaced it with a MC33079 from On Semiconductor. I was not
expecting much of a difference any more, but the difference
made by this "last piece of the puzzle" is quite astounding.

The MC33079 is a very good IC. It has low distortion (with
published distortion figures correlated with frequency and
output voltage). It has an output stage which doesn't
exhibit crossover distortion and is quiet. It is also stable,
being internally compensated for stability at unity gain
with, the data sheet claims, a good phase margin. It's not
a bad match for equipment loaded with NE5532's.
On Semiconductor, a 1999 spinoff company of Motorola parentage,
recommends this part to designers for audio work, along with their
version of the NE5532 dual op-amp and the NE5534 single.

The old chip in the MP-1's EQ section, the LF347, it a ho-hum
part whose datasheet does not recommend it for audio at all.
It is for DC applications only. No graphs of any kind are
given on the datasheet, which is kept very brief.
"Wide bandwidth" is claimed, but it is not stated how wide,
or whether it is unity-gain stable. "Low distortion" is also
claimed, but no figures or graphs are given. I suspect that
this op-amp quite possibly oscillates unless it is carefully
compensated with external components, and undoubtedly introduces
distortion, which has contributed in a big way to my MP-1's
sound quality issues. The LF347 should never have been used
in any quality audio equipment, and would never be used for such
a purpose by any competent designer today.

After I popped in the MC33079 I was surprised with a sound
which has become much more transparent and pure on all tones,
from solid state and tube cleans, to blues crunches to rock
and roll and metal distortions. The best aspect of this new
sound is that there is more musical high end. Previously,
I always struggled with high end. When I dialed in more, there
was something objectionable in the sound, and when I tried to
cut that away, the sound became too dull and blanketed.
Now the "fizz frequencies" sound more pleasant. I'm hearing
what is sometimes calld "bacon frying sizzle", and the high
end is more controllable. I'm guessing that this is because
it doesn't contain nasty distortion products that the
LF347 op-amps were introducing.

Basically, the rig sounds like simple tube amp, except
that it is MIDI-controlled. I.e. this is what it was supposed
to be like, until they screwed it on the assembly line with
sub-par op-amps to save money.

It is poor op amps which cause the MP-1 to be tonally regarded
by some guitarists as a preamp only for "80's hair metal".
Replace all the audio path op-amps with high fidelity units,
and you're in for a big surprise. It literally becomes a
different animal, tonally. And a very quiet one.

You know, if Vito Bratta had had an MP-1 like this,
he might have tossed the BBE "sonic minimizer" in the trash.