Eq for both bass and guitar?

[ghiekorg]

Hi everyone.
I am using the same pedalboard for bass and guitar. I am pretty happy with it as i only needed the OD/dist pedals to be different but i am sharing all the rest. I am just not completely happy with the EQ. Atm i am using a Boss GEB7, as i already had it, but the jump in the mid/high range is huge (the last 3 bands are 850/4.5k/10k).
Is there something that could be used with both instruments and could fit into a 1590B (maybe a 125B but i am not sure)? I couldn't find something specifically for it.
Thank you for your help

[Mark Hammer]

Really, the simplest and most flexible path would be a 3-band semi-parametric, that would let you use the same pedal and optimize it for bass OR guitar.  The Intersound IVP, discussed in this thread - https://www.diystompboxes.com/smfforum/index.php?topic=125723.0 - was the epitome of flexibility in the late '70s.  Part of that is because it has 4 semi-parametric resonant cut/boost sections.



This PAiA circuit also has 4 boost/cut sections.  https://www.experimentalistsanonymous.com/diy/Schematics/Tone%20Control%20and%20EQs/Four%20Band%20EQ.pdf

When a company makes an EQ pedal, they either target it specifically for a certain instrument, or else, having no idea what the end-user will apply it to, tries to cover all bases.  That's why it also includes a useless 10khz control.

In theory, an EQ pedal optimized for bass would simply move everything down by an octave or so.  Instead of a 200hz control, it would be 100hz, and so on.  The problem is that using a toggle to change 2 cap values for each band requires a lot of wiring and probably wouldn't fit in the existing enclosure.  The simpler solution is to build 3 bands, like in the Intersound unit, where a single pot allows you to alter the frequency where the boost/cut happens.  Three bands is pretty well enough to handle guitar and bass.  It is rare that adjustments are required in more than 3 bands, and for that we have bass and treble controls on the amp.

[iainpunk]

what is it you use your EQ for? maybe a parametric eq could be useful? or a few of them in a box, all having different ranges of course. that allows you to dial in the eq very precisely.

i have heard good things about this kit:
https://www.musikding.de/3-Band-parametric-EQ-pedal-kit

its a bit bigger, but its really powerful and precise.

cheers, Iain

[ghiekorg]

Thank you guys for your reply.
@Mark Hammer: Hi once again Mark i tried to follow the link of the intersound IVP but i think i would be too complicated for me to build. I have to still look at it properly though. 
The PAiA seems much easier and i could probably be able to make a pcb layout myself. I just don't get if it needs 12V to work. I only have 9V inputs in my supply

@iainpunk: that's a cool idea, but 10 knobs are a bit too much i guess... i will give it a look, maybe there is a way to avoid some knobs keeping the Q fixed for example. tomorrow i will check it when i wake up

thank you guys

[DIY Bass]

I have 2 Maxon bass EQs on my board.  I am pretty sure that Maxon also made them for Ibanez.  They also made a guitar version.  The centres are pretty well laid out for both bass and guitar I think.  The bass version goes from 50Hz up to 8kHz with pretty even spacing.  One thing about the Boss version from memory is that some of the mids frequencies which are so important for bass have weird large gaps between them.  I think it would be difficult to make a graphic EQ just because of making all those evenly spaced slots in the box, but the schematic is out there.  I think I would probably just try to find one second hand though.

[ThermionicScott]

I'll echo Iain's question, and extend it in all directions:  what all does this EQ need to do?  Do you have pictures (or a description) of how you currently set up your GEB7 for guitar vs bass?  Would you actually plan to crouch down and push a lot of faders around when switching instruments during a show, or do you have a good idea which bumps and dips you would always want for each?

I wonder if there might be a much simpler solution than trying to figure out how to shift a 10-band parametric EQ up and down on the fly... 

[ElectricDruid]

I wonder if there might be a much simpler solution than trying to figure out how to shift a 10-band parametric EQ up and down on the fly.

Yeah, this is an interesting idea. If you're using a particular setting for guitar and a particular setting for bass, could you build two fixed EQ stages and just switch between them? Or if "totally fixed" is just too rigid (and it is a bit rigid) then perhaps you could only go for the essential knobs on each one, thereby keeping the circuits simple enough to not be a total nightmare.

The thought here is that it's often much simpler to build one thing to do one very specific job, rather than build a general purpose tool that does many jobs (like "Multi-band EQ for both Bass guitar and guitar that does whatever").

[Rob Strand]

This one is very simple and performs well.   Used in many bass amps.

As shown +/-15dB and Q = 1.5.

If you want +/-12dB instead of +/-15dB, change the 1.57 in the calculations to 1.23.  For 12dB Q = 1.2.



It is also possible to raise the Q of the circuit by adding a resistor (see R16 on the middle circuit, which sets the max boost/cut).   Notice in this case  the frequency is f0 = 1/(2*pi*RC) and C1 = C2 = 1/(2*pi*R*f0).   You can also change the ratio of the caps and add R16. Normally the circuit sounds fine as is.



There was thread on this forum showing how to make a single supply version of the circuit.   To add more bands you just cascade as many bands as you need and change the caps for different center frequencies.

Maybe this one,
https://www.diystompboxes.com/smfforum/index.php?topic=95483

[ghiekorg]

I'll echo Iain's question, and extend it in all directions:  what all does this EQ need to do?

I am sorry, i completely overlooked the question while reading. And you're right, i don't need that much of a control. I actually use it mainly for 2 things: boost some bass frequencies (and cut some mid/high ones) when playing the bass in my dub-reggae band, and i use it to boost some mids/highs when i play guitar to give more brightness/cleareance/cut to the sound.

The thought here is that it's often much simpler to build one thing to do one very specific job, rather than build a general purpose tool that does many jobs (like "Multi-band EQ for both Bass guitar and guitar that does whatever").

That's actually my problem with the geb7. It's too complex. I mean, there are too many possibilities and often i end up with a worse sound that before

This one is very simple and performs well.   Used in many bass amps.
[...]
Maybe this one,
https://www.diystompboxes.com/smfforum/index.php?topic=95483

Thank you Rob Strand. That's actually a cool idea. I am a complete beginner so i don't understand much, though. Or i understand but i wouldn't be able to modify it  to my needs. I will try
I think probably for me would be enough to have two of them together. A 5 knobs version: bass freq, bass boost, treble freq, treble boost, volume.

thanks everyone

[ghiekorg]

I had a better look to the threads and the simplest way seems the one using George Giblet schematic. Problem is: i can't understand much.
I looked at this thread https://www.diystompboxes.com/smfforum/index.php?topic=95483 and i made a sketch of a schematic, based on it. It's just the more a read the more i have doubts (probably this whole thing is way above my capabilities, still let's see where it brings me).

Here is the schematic i made. My idea is to have 2 knobs controlling the frequences and 2 knobs controlling the boost/cut on them. I then wanna have a general volume control (i didn't yet draw it)


And here are my questions:
1- As i understood the red part is justa  buffer. is it right?
2- Somewhere in the thread the author says the EQ needs a booster before it. Can i just add a whatever booster and make its OUT eq's IN (i always think of the JHS minibomb as it has just a handful of components...)? and adjust probably a cap and a diode
3- If i add a booster before it i could use the booster level as a general volume i guess and avoid a volume control before the eq's OUT. Right?
4- The purple part is just converting the 9v to 4.5v, right?
5- What does the orange part actually do? Is it just another buffer?
6- Here is a big hole: I left the R and C values blank inside the green and blue parts (eq sections) because the formulas are not really clear to me...
C=1/(2*pi*f*R). I guess pi is 3.14 and R is the value of the resistors (is it?). Is f the frequence? But how shall i use the formula if i have a range of frequencies? I was thinking it would be good to have one knob going from, more or less, 50 to 800Hz and the other one going from 800 to 6400kHz.
7- Is Q 1.5 more or less an octave width?
8- Can i use 2x TL072?

Probably i have other questions but i guess for now is more than enough.

Thank you for your help

EDit: I am actually thinking if it would be just easier to take the ROG tonemender (http://www.runoffgroove.com/tonemender.html) and just modify it.

[ElectricDruid]

And here are my questions:
1- As i understood the red part is justa  buffer. is it right?

Yes.

2- Somewhere in the thread the author says the EQ needs a booster before it. Can i just add a whatever booster and make its OUT eq's IN (i always think of the JHS minibomb as it has just a handful of components...)? and adjust probably a cap and a diode

You can add a couple of resistors to that buffer stage and make it into a non-inverting op-amp gain stage instead (e.g. a booster).

3- If i add a booster before it i could use the booster level as a general volume i guess and avoid a volume control before the eq's OUT. Right?

You *could*, but with a non-inverting amp like I just suggested, you can't get less than unity gain (so you can only boost, not cut the volume).

But the main reason is that if you have the volume control at the end of the circuit, when you turn it down, you turn down the noise too. If it's at the beginning, you turn down the *signal* but any noise coming out of the circuit is still there.

4- The purple part is just converting the 9v to 4.5v, right?

Yes. And you might as well use that 4.5V supply rather than make another one with a pair of 2M2 resistors for the input buffer stage. Have a look at the first two op-amp examples on this page:

http://www.muzique.com/lab/buffers.htm

The second one is what you've got, with the two large value resistors (1M on the AMZ schematic, 2M2 in your case) but the first example is what you can do instead (Vr is your 4.5V bias supply).

5- What does the orange part actually do? Is it just another buffer?

Yes, but this one's inverting.

6- Here is a big hole: I left the R and C values blank inside the green and blue parts (eq sections) because the formulas are not really clear to me...
C=1/(2*pi*f*R). I guess pi is 3.14 and R is the value of the resistors (is it?). Is f the frequence? But how shall i use the formula if i have a range of frequencies? I was thinking it would be good to have one knob going from, more or less, 50 to 800Hz and the other one going from 800 to 6400kHz.

Yes, F is the frequency. Since you can vary R (with a pot) but you can't vary C, you be better re-arranging the formula to give you R, and then put your values in:

R = 1/(2*pi*f*C)

R min value = 1/(2*pi*6400*C)

R max value = 1/(2*pi*800*C)

Now you know what range your pot needs to cover. The minimum value is what you get when the pot is all the way at one end, e.g. the value of just R3 or R4.

7- Is Q 1.5 more or less an octave width?

I don't know. Here's a calculator that can tell you:

http://www.muzique.com/lab/gyrator.htm

(Yes! It's extremely close to an octave).

8- Can i use 2x TL072?

I would! That's the chip I use for *everything* unless there's a reason not to (Occasionally there is).

Thank you for your help

You're welcome. I hope it does help.

[antonis]

5- What does the orange part actually do? Is it just another buffer?

Yes, but this one's inverting.

For some reason hardly understood..

[ghiekorg]

That was SUPER helpful. Thanks a lot
I think i understood everything a bit better now.

Here is my new schematic, in red the changes i did. I actually  inverted the values for bass and treble resistor and caps. will fix it in the next version



Here is a list of the changes:
- Added a 2 resistors to make it a booster. I just can't find an easy way to calculate how big they have to be.
- Added a volume pot at the end (100kA would be enough? i actually have a lot of 250 and 500k A pots i am not using)
- Removed the 2x 2.2M resistors and used a 1M coming from the 4.5v
- I calculated the pot's resistance. I left the other values as they were in the example, both caps and resitors (apart from R3, R4, R7 and R8 and the related pots). Here is a google sheets file if you wanna check https://docs.google.com/spreadsheets/d/1pjZK0ikUGP-hXr9vIT-1fwqr47efZNNU7o8L5LKKoIc/edit?usp=sharing
And that's what i did:

. I first set the Cs so they would match the example and i found that a C of 7.8nF gives me C1 and C2 of circa 4.7nF and 12nF.
. In the example the frequencies are 200 to 2000Hz so i used the formula R = 1/(2*pi*f*C) to calculate the Rs and i actually got 10k and 100k, which seems to fit
. I changed the data and i found the values for 50, 800 and 6400 Hz which are 3k, 25k and 400k so i was thinking of using 3k resistors and 25k pot for the treble and 22k and 500k pot for the bass (if you have suggestions about the Hz values, feel free)

I am not sure i have to leave the second output buffer as it is (actually antonis wrote while i was writing this)

Thank you 

[ElectricDruid]

5- What does the orange part actually do? Is it just another buffer?

Yes, but this one's inverting.

For some reason hardly understood..

Lol, yeah, I agree, but I decided to gloss that over...until you came along!!

[ElectricDruid]

Here is my new schematic, in red the changes i did. I actually  inverted the values for bass and treble resistor and caps. will fix it in the next version

Hey, nice work! This is really getting there!

- Added a 2 resistors to make it a booster. I just can't find an easy way to calculate how big they have to be.

It's the standard non-inverting op-amp gain equation. If "Ra" is the one at the top, and "Rb" is the one at the bottom, it's:

gain = (Ra/Rb) +1

If you want to know what a gain of X is in dB, this is useful:

http://www.muzique.com/schem/gain.htm

(AMZ tools ftw, again!)s
Sometimes I think about it as "x4" or whatever and sometimes as "+12dB". Depends on the situation.

- Added a volume pot at the end (100kA would be enough? i actually have a lot of 250 and 500k A pots i am not using)

Yes, that looks good. 250K and 500K pots will start to present more loading to following circuits, so you'll get some loss but it's not mind-bindingly serious. I think they'd be fine. Imagine if you plug into a pedal with 1M input impedance, and you have a 500K pot set at 50% volume. The bottom half of the pot is in parallel with the 1M input impedance, so 250K || 1M = 200K. The top half is still 250K, so your 250/250 split has just become a 250/200 split - e.g. it dropped from 50% to 44%. Not unbearable, but worth being aware of.

- Removed the 2x 2.2M resistors and used a 1M coming from the 4.5v

Excellent. Note that because the impedance of the original circuit is the 2M2 resistors *in parallel*, you don't lose any impedance by doing this (or not enough to matter: 1.1M = 1M, near enough).

- I calculated the pot's resistance. I left the other values as they were in the example, both caps and resitors (apart from R3, R4, R7 and R8 and the related pots). Here is a google sheets file if you wanna check https://docs.google.com/spreadsheets/d/1pjZK0ikUGP-hXr9vIT-1fwqr47efZNNU7o8L5LKKoIc/edit?usp=sharing
And that's what i did:

. I first set the Cs so they would match the example and i found that a C of 7.8nF gives me C1 and C2 of circa 4.7nF and 12nF.
. In the example the frequencies are 200 to 2000Hz so i used the formula R = 1/(2*pi*f*C) to calculate the Rs and i actually got 10k and 100k, which seems to fit

Ok! That's really good that your calculations match the original values. That's the kind of confirmation that we're on the right lines that we're after! It doesn't happen all the time...hohoho...

I changed the data and i found the values for 50, 800 and 6400 Hz which are 3k, 25k and 400k so i was thinking of using 3k resistors and 25k pot for the treble and 22k and 500k pot for the bass (if you have suggestions about the Hz values, feel free)

There's a good page about this type of EQ you should have a look at:

https://sound-au.com/project150.htm

Rod Elliot reckons you can get about a decade of adjustment out of this circuit, so the "minimum value" resistor (R3/R4) should be 1/10th of the value of the pot. He doesn't say why, but I've learned to trust his experience, so I'm willing to accept that there is a reason, even if not explicitly stated. I notice the original values you mentioned for the circuit fit that pattern: 10K and 100K.

That would mean that your 800/6400Hz values are fine and should be possible. The decade would be 640 to 6400Hz, so you're on the safe side, which is a good place to be. It does mean there's something weird going on with your 22K/500K bass values. 3K/22K is ok for the Treble (2K2/22K would have been Rod's suggestion, so you've backed the range off a touch, which follows 800/6400 not being quite "x10"). but 22K/500K is "x22" so something is fishy there. What did you do differently from the first calculation?

I am not sure i have to leave the second output buffer as it is (actually antonis wrote while i was writing this)

No, neither Antonis or I can see any reason for why they've done that buffer as an inverting stage rather than a non-inverting one. It may have made sense in some previous earlier version of the circuit and is just a hangover from then. There's certainly no *advantage* to it here. My inclination would be to swap it for a non-inverting one like you have at the front, but without the gain. Unless you really like LOUD, of course

Edit: Sorry, without the gain, not with the gain.

[iainpunk]

i think that you don't need that output buffer, since the 2nd tone circuit has a 'strong' output. i'd use that opamp for a Vbias buffer because the tone circuits work better/cleaner with a stronger 4.5V point

cheers, Iain

[antonis]

IMHO, last stage buffer WITH gain (x2 or so) should make Volume pot more functional..
(just to oppose Iain..)

[Rob Strand]

You only need the inverting stage if the number of EQ bands is odd and the EQ stage inverts the signal, which is the case for the George Giblet EQ.

Here's a good pattern to follow.   Don't forget to connect the EQ bands to VREF as indicated on the GG EQ as "single supply".

[iainpunk]

IMHO, last stage buffer WITH gain (x2 or so) should make Volume pot more functional..
(just to oppose Iain..)

ill one up you here:

the input op amp should have a gain of 2 or 3 and a volume control after it, so you can adjust that to not clip so you don't need that output op amp's extra gain, so you can still free it up to buffer the Vbias. 


cheers, Iain

[ghiekorg]

Hey, nice work! This is really getting there!

Thank you ElectricDruid

gain = (Ra/Rb) +1

So as 12dB=x4 gain then
4=(Ra/Rb)+1
3=Ra/Rb
assuming Ra=10k then
3*Rb=10k then Rb=3.3k

Right?

There's a good page about this type of EQ you should have a look at:
https://sound-au.com/project150.htm

That's quite complicated for me actually but i tried to follow it and understand what i was reading
He seems to do the opposite of what i did: he changes the caps values keeping the R to 10k and the pot 100k.
Problem is: it doesn't match my calculations...
Example: his chart shows that, if we use both C as 22nF cap, f=1/(2*pi*R*C) should give me the a frequence of around 700 using a 100k resistor but i get 70 [1/(2*3.14*100000*0.000000022)]
I mean, bigger caps=lower freq. If he used a C of 7.8nF to have a 200-2000Hz band then it's quite weird to me that using a larger 22nF i get 700-7000Hz (which is of course higher)
In my sheet i made an interactive chart on the lower left:



As you can see it is in line with the 200-2000hz example but not with the chart


About the buffers: Now i am confused x2/x3/x4 gain? do i need that at the end? ianinpunk's post seems the easiest solution to me at least i don't have to draw it myself

thank you guys