5 Band EQ Project

[coldcraft]

Here's something I've been working on for a while. I've built a few precursors, but this one has not been tested yet.

Its a 5 Band Parametric/Shelf EQ that mates 3 Bands of Gyrator Band Pass with a pseudo-baxandall High and Low Shelf.

The Parametric Sections are based on an RG Keen article at geofex. The Bax-ish stuff is adapted from what I've seen in a few overdrive designs.

The controls are as follows,
Bass Cut/Boost
(3x) Mid Cut/Boost - Frequency - Q
Treble Cut/Boost - Shelf

Treble Shelf control lets you use the high shelf as a high roll off. It does not however prevent the opposite action of reverse roll off, boosting higher and higher frequencies more and more.

Some nice aspects of the design,
All bands are tweakable by some simple capacitor choices. Change the range of the parametric bands and change the corner frequencies of the shelf bands. This makes the design tweakable for Bass, Electric or Acoustic.

The Shelf EQs allow for the possibility of asymmetrical Boost/Cut, meaning, Boosting occurs with 1 corner frequency, whereas cutting occurs with a different corner.

The gain range of the shelf bands and the parametric bands are set separately through simple resistor choices. More on this after I verify the design.

18v headroom.

Anyway, here's the design and a preliminary layout. I'm going for a horizontal 1590B/NS PC-mount type thing. The Frequency pots would be this 9mm board mount trimmer type (http://www.mouser.com/catalog/specsheets/TW-700180.pdf), with typical 16mm board mounts for the boost/cut. Comments, questions and concerns are appreciated.





One last thing. I was planning on offering etched PCBs of this project for a few bucks. I'm also considering producing fab'd PCBs if I'm happy with the design post-verification. Please shout out if you might be interested in one or the other. Not looking for a group buy, just trying to gauge how many PCBs I can afford to make.

[Silent Fly]

Interesting project Coldcraft.

To be honest,the power supply doesn't convince me. It looks a little bit convoluted.

Why don't you get the +9V and GND from the circuit power supply and -9V from the 1044? In this way you get a dual power supply and you can remove coupling capacitors and bias voltages.

I may be wrong but the Freq1 and Q1 pots of the Mid control section are in series. I doubt that one controls the frequency and the other the Q factor.

[coldcraft]

Frequency and Q work as they should. The parametric bands have been tested and the work really well actually. They do connect, but they are not in series and they are separated with a capacitor across the OpAmp Follower Inputs. RG has a great explanation of the Gyrator design on geofex.com

I hadn't considered a bipolar power supply because I've never used one before. How exactly would it simplify things?

How is the 18v supply convoluted? I believe I've set up the standard way of achieving 18v from the Max1044. Also, if I use a bipolar supply, is it possible to still use an opamp to create the Vref (now ground) as I've done here with 9v? Otherwise I'd have a wasted Op Amp half.

If you look at the schematic, there are actually only two coupling capacitors, the input and output. I suppose it would make it a bit simpler to drop 2 caps and 2 resistors, but it wouldn't really make a difference in the layout. Also, I would like to make the PCB such that the 18v headroom is optional, and that 9v could be used quite easily. This would mean a few added spots for parts that aren't on the schematic yet.

[coldcraft]

well, doesn't seem like there's too much interest. I probably won't tie up cash in a Fab'd order just yet. Once I verify it, I will have etched boards available if anyone is interested in that.

[soulsonic]

I recommend using a 9v zener diode instead of 1N4001 for the reverse polarity protection. You want to protect the 1044 from reverse voltage, over-voltage, and ESD. A zener takes care of all of that.

Also, if you were doing it as with a bipolar supply, you could just break it up into a quad opamp for the gyrators and tone amplifier, and then use a single for the input buffer. I would recommend a JFET input like the LF411 or TL071 for the input buffer and use 1M for R2. That should give lower input noise from the lower currents. Just a couple suggestions, the design looks pretty good as is.

[coldcraft]

Hey Soulsonic, thanks for the feedback/suggestions! Zener sounds like an excellent idea.

Interesting suggestion to move to a quad opamp. I'll have to see how that changes the layout, but I'm all for lower noise. The schematic is actually backwards a bit, I was planning to use TL072s in the audio path and 5532s for the Gyrators.

[coldcraft]

here is the latest layout. this is not verified in full yet.

[livingnote]

Haw cool! That would be a nice thing to stick out the other side of my compression sustainer.

[coldcraft]

Haw cool! That would be a nice thing to stick out the other side of my compression sustainer.

just read your PCB thread on the DIY forum. Any chance you have a tutorial or tips lying around on how you do such beautiful etches?

[livingnote]

Sure - I have my old etching page here and some general stuff about Illustrator for PCB.

Thing is - I really need a tone control for my stompbox pile, EQing around in Cubase is just not "it"

[spongebob1981]

This REALLY catched my eye
Though my english is far from great, didnt understand if this was to be a stompbox or a rack unit.
Is it still in development or stopped? usable yet?

thanks and see ya!

[coldcraft]

This project is still in development. I guess it should have been first posted in a different forum area. I've revised it and tweaked it, and at the moment I'm busy with other projects. I would like to wait until I've finalized it and verified it myself before releasing a new version.

Perhaps bassists don't agree with me, but I feel 5 bands is overkill for any guitar application. I"m toying around with the idea of providing both, 3, 4 and 5 Bands version where the only difference is the number of parametric mid bands. Each version will have Bass/Treble Shelf EQ in common, but the corner frequencies will be different.

[spongebob1981]

I agree with you. Personally I think I should make it work with 2 or 3 parametric bands and no other bands (baxandall, shelf, w/e)
Plus, I have some metal boxes I'd like to use to make the stompbox and no more than 6 knobs will fit
And many fellow bass players agreed with me on this. A 9v powered, 2 band parametric should be tha thing.

As I said before: keeping an eye on this one
best of lucks

[coldcraft]

i think the only problem I can see with only two parametric bands is that the band range is only about 4.5:1, so from 100hz to 450Hz, etc... Do you see that as a problem in your application?

[spongebob1981]

I'm shooting for use it as a bass guitar stompbox. So perhaps it is a problem.

Best case scenario: 2 or 3 bands (better 2, for simplicity and knob count sake)
40hz - 5khz range across the mentioned bands.
9v operated (battery or external power supply, w/e)

Compromise scenario:
60hz - 3khz range

Anything more than that just deserves to be a rack unit (or even as part of a pre-amp) also posible, but not as a simple solution as a stompbox could be.
(But entirely doable, mind you)

So, if your project can be thought as this, great.
Otherwise, keep on the track you planned, I might be able to use it either way XD
good luck!

[cAMPus]

Hi,

verified project: http://www.musiker-board.de/faq-worksho ... mpbox.html (sry, it's in German but ...)

+-9V was to less to work without distortion for a active bass at maximum boost of the eq bands.

That's why I used a LT1054 for bipolar voltage doubling from +9V to +/-~17V.

U.

[spongebob1981]

I can't read german but that link of yours is quite impressive
I'll check if the opamps are available in town and give it a try. (Google translator I summon you!)
see ya!