Custom EQ

[kevilay]

Hey guys I was interested in building an EQ pedal. Anyone have any good schematics? Also where can I get those little slider pots? What is different between a guitar eq and a vocal eq?

Thanks
Kevin

[Seven64]

http://buildyourownclone.com/eqinstructions.pdf

[Mark Hammer]

Slider pots are nice and compact, but they are an absolute bugger to machine for. Moreover, they are more susceptible to dirt getting in than rotary pots are, particularly given that they are sitting on the ground face up.  A better choice are some of the small PCB-mount pots like you see on Z-Vex, Way Huge, or Fulltone pedal.  They can be gotten from Mouser and similar suppliers, and you can stick a whole whack of them in a small chassis.

"Graphic" EQs that use sliders are intended to be general purpose devices.  The reason why they have so many bands is because the manufacturer has no idea what bands will be important to you.  The reason why they are sliders is so that one can instantly get a sort of mental frequency plot by looking at the slider knobs.

If you know in advance what bands will be most critical to the task at hand, you can usually get away with fewer bands.  You will note that parametric equalizers, which allow you to dial in the frequency band of interest, rarely have more than 3-4 bands of adjustment.  If you know what regions of the spectrum will be important to what you want to do, 4-5 bands of gyrator curcuit, like the BYOC schematic linked to, will do you nicely.

[JRM]

If you realy want a graphic EQ it's cheaper/easier to buy an used one and then mod it or rip it off completly and install all new circuitry.

[kevilay]

if I want it for a guitar maybe I should go with something else? like a 3nob eq? What do you guys think?

[JRM]

Why just 3 knob? You still can have a 100/200/400/800/1600/3200/6400Hz EQ with standard rotary pots. If I build one EQ it will probably be like that. Check GGG or other and you'll find it.

[Mark Hammer]

Said it before and I'll say it again.  A decent flexible EQ unit for guitar would have a Baxandall-type bass shelving section, two semi-parametric resonant section with adjustable centre frequency and cut/boost, and a tunable 2-pole lowpass filter, with perhaps optional variable Q for getting a bit of peak at the corner frequency.  Six or seven knobs would get you an enormous amount of flexibility.

[JRM]

Said it before and I'll say it again.  A decent flexible EQ unit for guitar would have a Baxandall-type bass shelving section, two semi-parametric resonant section with adjustable centre frequency and cut/boost, and a tunable 2-pole lowpass filter, with perhaps optional variable Q for getting a bit of peak at the corner frequency.  Six or seven knobs would get you an enormous amount of flexibility.

That sounds VERY interesting Mark! Is there anything like that already developed?

[Mark Hammer]

Well there are certainly the component circuits available, though I am unaware of a single circuit and board layout that collects them all together.

But consider a bass shelving control like that shown here: http://freecircuitdiagram.com/2008/09/02/baxandall-tone-control-using-lm833-thcircuits-schematic-diagram-and-the-formula/ 

....and two tunable resonant sections like those shown here:  http://www.geofex.com/article_folders/eqs/paramet.htm

...followed by a two-pole (12db/oct) lowpass like this one:  http://www.radio-electronics.com/info/circuits/opamp_low_pass_filter/op_amp_lowpassfilter.php

What counts as "bass" for guitar is reasonably constant enough that having a fixed bass boost/cut control (i.e., "fixed" in the sense of pivoting around one corner frequency) is fine. 

Two semi-parametric resonant sections (i.e., cut/boost and sweepable centre frequency for each), such as RG shows, can cover a broad range of frequencies.  The Q, or selectivity can be fixed to avoid the hassle of needing to find room for extra controls, doing the wiring, etc.  Probably easiest if one installs individual machined socket pins and treats the "resonance" control as a fixed resistor, but tries out different fixed resistor values in the socket.  My own preference would be to have a slightly higher resonance (greater selectivity/focus) for the lower resonant band, and slightly lower resonance for the upper band. 

One section with variable centre frequency from 200hz-2000hz, and a second section with centre-frequency range of 400hz-4000hz would provide ample coverage of regions of interest with enough overlap to achieve complex tonal variations.

The lowpass filter could be sweepable over a couple of octaves(say, between 2khz and 8khz), which would allow for bright glassy single coil pickups to sound "rounder" and more HB-like or P90-like.  It would require a dual-ganged pot to work, but one could also use a DPDT on-off-on toggle to switch resistors or caps for 3 fixed lowpass rolloff points if you wanted to keep things simple.

Presumably, all of this complements what you have on your amop.  Between the 6 controls of this hypothetical pedal circuit, and what's on the amp, there is an extremely borad range of guitar-relevant applications, and very little limitation on what you could achieve

For example, want a scoop mid?  Crank the bass, dial in the optimal midpoint with the broader-range resonant section and cut, set the lowpass filter around 5-6khz, and use the narrower-range resonant section to add just a touch of bite near the top of its range.  And so on.and

[JRM]

I realy can't disperse my attention to other projects before I finish my multi-fx unit but that will be a project that I might consider in the near future: gatter all the 3 circuits and put it in one single board...

[kevilay]

mark, you sir are a very smart person. An eq is much more complex then I imagined it. If you throw a circuit together ill breadboard it, and if all is well I will put it into a pcb board layout on eaglecad.

[gregwbush]

Here's an example of a semi parametric EQ. Has some maths on the schematic for selecting components for desired frequency range and that.

http://s476.photobucket.com/albums/rr130/ggeffects/?action=view&current=gg_simple_parametric.png

Just on it's own (with an input buffer added), it's a good example of how versatile a parametric EQ section really is
My experience building one... http://www.diystompboxes.com/smfforum/index.php?topic=95483.0

I can imagine how versatile 2 of these (or something similar) would be, PLUS the baxy bass n treb. Very versitile indeed

[Mark Hammer]

Personally, I think design of a good EQ unit for guitar begins by asking what it is one wants and needs such a unit to do for a guitar.  The second step then involves pondering what is easily, or more effectively taken care of by the broader context, including the controls on the amp, the guitars themselves, and any other pedals on the way.  Once you have the "problem" that needs to be solved clearly defined, solving it is a breeze.

[gregwbush]

As a half baked hack, a copy and paste-er, a mesh together and hope for the best, electronics guess-alot...

I realize how simple your "equation" is Mark, and at the same time; how far away i am from solving it. What's the answer to a question i haven't yet defined?! Null!

Anyway this isn't my thread

[Mark Hammer]

One of the things I've learned in all the organizations I've been employed in or affiliated with is that if you have what folks consider a "good idea" even 3 nanoseconds before someone else comes up with the identical idea, it is frmly affixed to yur lap, and no one else will EVER take steps to see it through.  I think it's going to be up to me to put my design skills where my mouth is and treat this as a job duty.  I'll get on it.

[DavenPaget]

One of the things I've learned in all the organizations I've been employed in or affiliated with is that if you have what folks consider a "good idea" even 3 nanoseconds before someone else comes up with the identical idea, it is frmly affixed to yur lap, and no one else will EVER take steps to see it through.  I think it's going to be up to me to put my design skills where my mouth is and treat this as a job duty.  I'll get on it.

Well , what do we say when we give a +10 band rating ?

[kevilay]

dont feel obligated to design anything. I am going to play around with his example uptop. Mark your advice is great!

[Ben N]

dont feel obligated to design anything. I am going to play around with his example uptop. Mark your advice is great!

Bite your tongue! Design, Mark, design!

[JRM]

Although I shuldn't (see my previous post) I've been thinking about this. The scale of a 24 frets standard tunning guitar ranges from 82,4Hz (6th open string E2) to 1318.5Hz (1st string on the 24th fret E6). 2nd Harmonics range from 165-2637Hz, 3rd harmonics between 247-3956Hz, 4th harmonics 330-5274Hz and 5th harmonics 412-6593Hz.
I think this might be useful for the designing limits. As one can easily see, in the usual 100/200/.../1600/3200/6400 graphic eqs the last three bands affect mainly/only harmonics.
I've tried to calculate a Baxandall Bass shelving with a lower turnover frequency on E2 (82,4Hz) and a +/-3dB frequency on E4 (330Hz). If my calculations are correct, one can achieve it with a 25k resistance in series with a 100k pot in parallel with a 19nF cap. I wish I could see the response of that now but I'm at work...

[gregwbush]

Okay....

I decided to make a start on this custom EQ thing. Here's what i have so far...