ampage Tube Amps / Music Electronics	For current discussions, please visit Music Electronics Forum.

ampage archive

Vintage threads from the first ten years

DIY Parametric EQ at GEO

:

10/6/1999 1:00 AM
R.G.	DIY Parametric EQ at GEO I just posted a DIY design for a two-band parametric EQ at GEO. This thing uses two dual opamps and gives two bands of EQ, with about 12db of boost/cut and variable frequency and Q per channel. It's gyrator based, by the way.

10/6/1999 10:56 AM

Note: the simple "graphic EQ" shown in the first schematic is exactly the same topology used in the MESA/Boogie amps, only the MESAs use a "discrete op-amp" (three transistors) instead of U1B. So if one wishes to build an inductor-less EQ for a boogie clone go and read RG's article.

Also note that the simulated inductors are built around an op-amp wired as an unity gain amp. You don't need an op-amp for this: you can use a commom-collector transistor stage (the Boss Metalzone uses this trick).

10/6/1999 12:49 PM

Yep, all correct. The LC filter style graphic EQ is widely used, both in hifi equipment and in a few amps. And any high input impedance, low output impedance voltage follower will work for the "gyrator". It's hard to get enough performance out of a single transistor for this, but two-transistor connections of various styles will also work as well as an opamp, especially a FET-transistor pair. A single transistor follower causes a highly idiosyncratic sound, depends a lot on the transistor doing the following.

10/6/1999 2:47 PM

I guess you mean a single transistor follower has different sink/source capabilities (and output impedances) and a tendency to oscillate when driving a capacitive load.

Don't these single transistor drawbacks only show if you're dealing with a high level sound (such as just before the power amp in a boogie) or do you have to bother with it in a low-level situation (fx pedal)? Aren't those "idiosyncracies" of the single transistor follower mostly related to large signal operation?

10/6/1999 3:50 PM

Yes. The single transistor usually has too low an input impedance - in this instance it shouldn't noticeably load a 1M resistor - and as you mention, that pesky tendency to oscillate. A two-transistor rig can fix the input impedance as well as being much more linear than a single transistor, and may even have some advantages over an opamp in terms of gain-bandwidth for gyrators above a few kHz. However, unless you're already running into these problems, a two transistor rig will use more components and board space than an opamp, so it's something that should only be used where it's really needed.

The drawbacks to a single transistor do show up even at low signal levels. It's worth while to note that nonlinearities and quirks are sometimes very useful musically, so it's worth messing with them to see how they sound. I suspect that a musically useful quirk from a transistor would depend on transistor properties, like the sound of a Fuzz Face does.

In any case, the right thing to do is to experiment!!

10/6/1999 3:56 PM
GFR	Thanks!

10/7/1999 3:17 AM

R.G.
Very nice and timely.
but I'm really struggling with the math.
you mention Resonant Fr as 1/6.28*SQRT(L*C)
SQRT, is that square root? Is the series
resistor ignored?
Later you mention Center Fr as 1/6.28*L*C
I suppose to an EE, this is childs play,
but I'm stumped. For those of us in the
peanut gallery, could you maybe give an example
with "real" numbers. For example, in the common
as ticks on a dog, TSx - the 1k/.22 to ground,
I added a 56mh. choke in series between the .22
and ground. Whoa-instant deep notch at
about 1.2hz, and great highs from there up.
But I can't make the math work. Help please.
Thanks Ed Rembold

Page 1 of 2