Utilizing George Giblet's parametric EQ ...? Schematics included

Started by gregwbush, January 09, 2012, 04:07:17 AM

Previous topic - Next topic

gregwbush

Hello all

I've decided i want a stand alone mid boost/cut pedal, 2 knobs; Freq. and Mid.

I found this...



What i'm wondering is what to put before and after...??

An op-amp buffer in front, maybe?
And after; just a DC blocking cap?

I'm also not 100% on the Vcc/2 supply configuration. Will the resistor values i choose effect the behavior of the EQ?

Here's what i've come up with so far...

Will that work? What should i add or remove?

Excuse my ignorance
Any thoughts / ideas would be great
Edit my schematic for me!

Cheers,
gregwbush


gregwbush

Update:

I built the thing as in the schematic above and it seems to work!

B10k pot clearly cuts mids below 12 o'clock and boost mids above 12 o'clock
The 2 100k log pots i'm currently using (trying to keep them both in the same position of course when i adjust), clearly changes the frequency i'm boosting/cutting. It's late now so i can't really turn the volume up and test properly.

I'm still very curious whether the way i've got it rigged up makes it work "as expected"

Any ideas, thoughts?

Morocotopo

If it works, don´t fix it!

Joking aside, I´d replace the 2,2M resistors at the buffer opamp with a single 1M/470K one connected to the Vcc/2. Also, the two 100K´s I´d replace with two 10K. They have no effect on the sound, except for adding noise if they are too large. Also, you could replace the two 10uF caps with smaller ones,  1uf or so, that would allow you to use non electrolitic ones, those are generally considered not the best for audio, always checking that they don´t eat lows. These days you can get non polarized polyester ones of those values, the "greenies" A.K.A the chicklets, or the "blue boxy" ones.
I´m no expert, just things I saw on other schems.

EDIT: As it is, the thing inverts the phase of the signal. Since you have another opamp (the buffer), I´d configure that as an inverting buffer, so as to get at the output a non inverted signal. It might not be important, except if you use it in parallell with other FX´s.

EDIT 2: And, if you plan to use it as a stompbox, I´d add a 1M R to ground at the input before the cap and a 100K at the output to ground after the last cap.
Morocotopo

wavley

New and exciting innovations in current technology!

Bone is in the fingers.

EccoHollow Art & Sound

eccohollow.bandcamp.com

JRay


Here is a similar circuit that offers a sweep of a little over 100Hz to around 8Khz...

The dual reverse log pots are a sourcing problem in these circuits as they are like hen's teeth, but a standard dual log pot will do, it just means that your control will operate in reverse to what you would expect with the sweep going from high to low as you turn the knob clockwise. I have no idea of the Q of the circuit but it sounds good and is very useable. Hope it helps

     

Cheers
Ray

gregwbush

Thanks for the alternate circuit ray
About the antilog pots, unless you wire them backwards, they shouldn't respond backwards, just differently. Still bass at counter clockwise and treble at clockwise

http://sound.westhost.com/pots.htm
This article seems to define C labelled pots as the old school label for Log and F as antilog. So i'm under the impression that C is the same as A, and the pots in these circuits are log not antilog. Someone correct me if i'm wrong

gregwbush

Quote from: Morocotopo on January 09, 2012, 10:52:06 AM
If it works, don´t fix it!

Joking aside, I´d replace the 2,2M resistors at the buffer opamp with a single 1M/470K one connected to the Vcc/2. Also, the two 100K´s I´d replace with two 10K. They have no effect on the sound, except for adding noise if they are too large. Also, you could replace the two 10uF caps with smaller ones,  1uf or so, that would allow you to use non electrolitic ones, those are generally considered not the best for audio, always checking that they don´t eat lows. These days you can get non polarized polyester ones of those values, the "greenies" A.K.A the chicklets, or the "blue boxy" ones.
I´m no expert, just things I saw on other schems.

EDIT: As it is, the thing inverts the phase of the signal. Since you have another opamp (the buffer), I´d configure that as an inverting buffer, so as to get at the output a non inverted signal. It might not be important, except if you use it in parallell with other FX´s.

EDIT 2: And, if you plan to use it as a stompbox, I´d add a 1M R to ground at the input before the cap and a 100K at the output to ground after the last cap.

Thanks for your input Morocotopo!

Is this what you mean?



I haven't verified the above schematic yet, as boredom kicked in while waiting for replies on the matter so i brought out the perfboard and built it as is in the first post.



Replacing components on perf could be a pain. Might have to start again. Oh well, i enjoy layout-ing and soldering anyway

Talking of cap types, all i get these days are "greencaps" or the box type polyester ones shown above. And ceramic for pF values. That's what jaycar (local 'lecky suppliers) have, so that's what i have.

Anyway;
Morocotopo Edit 1: I don't know how to wire an inverting buffer... feel free to draw on my .PNG

Morocotopo  Edit 2: Good point. Too easy. One question though... how does that (and the swapping of 2M2/2M2 input bias thing for 1 Meg to Vcc/2) affect the input impedance and response of the circuit? It was high and now it's two 1 megs in parallel yeah? 500k

Obviously my theory aint so crash hot. I tend to copy bits and pieces and try to mesh them together, hoping for a result.

So keep the idea's coming people!

Cheers,
gregwbush

Oh, and any info on Log / Reverse Log and their labeling would be much appreciated. My brain says:
A = C = Log,
B = Lin,
F = Reverse/Anti Log

My brain could easily have been mislead by the interwebs though

Morocotopo

http://hyperphysics.phy-astr.gsu.edu/hbase/electronic/opampvar.html#c2

http://www.eng.yale.edu/ee-labs/morse/compo/sloa058.pdf

In there you have some explanations on how to make an inverting gain stage, that, if you set it up with a gain of 1, works as an inverting buffer. The second link is full of useful bits and pieces, to study and/or build.

Regarding the caps, either of those are OK, I get the yellow boxy ones in blue in my country.

I see in your pics a breadboard... set up the schem there and try the changes, so you can hear what, if any, effect they have on the thing. Once satisfied, make it on perf!

About the impedance, I´m not so sure, but anyway for electric guitar use 500K is OK mostly. Try both to see if there´s an audible difference.


Pots are usually A:log, B:lin, C:reverse log as far as I know. Never seen one marked F.
Morocotopo

gregwbush

Thanks for the links. Will have a look when i get a moment

It was pretty dumb of me to not breadboard this one. Turns out i drew it up as a perf layout on the fly, before drawing and posting the schematic here. Oh well

I punched c100k into ebay and most results do indeed say reverse log

gregwbush

Mmmkay...

Here is my VERIFIED schematic, which i find useful in front of an AMZ buffered minibooster. So much so that i made a board with both circuits (switchable EQ on/off)



I'm not sure of the range of this thing with the 50k linear Freq. pot... but it is a good range for mid boosting or cutting to my ears. I tried linear 100k and found that george's suggestion to limit the range to a factor of 4 or 5 for linear pots made the control have a better feel to it

cheers for now,
gregwbush

gregwbush

Here's my verified ParaEQMiniBoo perf layout



I took out the 1 Meg resistor in the minibooster and replaced it with a 1 Meg Log pot. I call this the GRIT control







The odd layout is so it fits around the foot switch!

gregwbush



Controls are
Freq.
Mid.
Grit
Level

Still gotta put in the EQ bypass switch. I'll make it fit somehow. This will also be a phase flip switch but that's fine with me~

cheers,
gregwbush


gregwbush

Thanks JRays!!!

gregwbush

Tight fit with 16mm pots!



Impossible with 24mm pots. The ones you see here are "dummy" pots until my order arrives

earthtonesaudio

Nice job Greg. 
There is a very simple solution to your EQ bypass problem that does not invert the polarity of the effect;

Simply a SPST switch between Vref and the EQ-section op-amp's non-inverting input.  Connecting the non-inverting input to Vref turns off the EQ but keeps the op-amp in the inverting mode.

Simulation link

gregwbush

Ha!

Thanks mate

Things seem so simple when someone else tells you how to do it lol

cheers
gregwbush

gregwbush

Quote from: earthtonesaudio on January 24, 2012, 10:46:23 AM

There is a very simple solution to your EQ bypass problem that does not invert the polarity of the effect;

Simply a SPST switch between Vref and the EQ-section op-amp's non-inverting input.  Connecting the non-inverting input to Vref turns off the EQ but keeps the op-amp in the inverting mode.


Ok

I had a good think about doing this and decided against it, instead using a DPDT toggle switch to "true bypass" the whole EQ section. The reason i did this is so in bypassed mode the pedal is a standard mini-booster, rather than a pre-buffered mini-booster (pre-double-buffered-inverting-mini-booster to be more specific)

Phase inversion and or on the fly flipping of phase isn't a problem for me. There's no scenario i can think of where an in phase signal will be mixed with an out of phase signal. Not at the moment anyway. Who knows what my setup might be like further down the track. I'll deal with it then!

Now, back on topic

I started this thread saying i wanted to build a 2 knob box with freq. and mid. controls

The following schematic (currently partially unverified) is a 2 knob job, does NOT invert phase, and uses the spare op-amp of a TL074 or pair of TL072's to light up CUT and BOOST LED's depending on the position of the cut/boost control. This concept is discussed and demonstrated (video) in a thread i started the other day: http://www.diystompboxes.com/smfforum/index.php?topic=95737.0



Anyone see anything wrong with the above schematic?
I dare say coupling caps could be smaller, but i don't know enough theory

Cheers,
gregwbush

slacker

Looks good to me, just a couple of small things I'd change. You don't need the 10uF caps after the first and second stages, they can be replaced by jumpers. If you do keep caps in those positions they shouldn't be polarised. The two 2M2 resistors could replacef by a single 1M resistor from the opamp's + input to 4.5 volts.

gregwbush

Ok thanks for those tips. So no cap's needed between stages? I'm in the process of doing a perf layout now, so i'll leave those out and change the last cap to 0.47 np. As for changing the 2m2's, because i've already got half of this layout layed out... they can stay. Will post layout soon and start building!