Variable crossover for bass

Started by Kummeli, December 12, 2016, 02:29:09 PM

Previous topic - Next topic

Kummeli

Hi from Finland!

New to this forum. Hope you guys can help me out straight away. :)

So, I would want to build a "splitter & mixer"-pedal with a variable crossover for our bass player. Idea would be that he could run the full range sound through boosters and octavers but could run just the higher frequencies (crossover variable from 100-800hz or something like that) through he's dirt pedals and modulation.

Basically something like this: http://www.sfxsound.co.uk/xm/

BUT, instead of 'LOW' and 'HIGH' loops there would be 'FULL RANGE' and 'HIGH'-loops. Or maybe even better if the 'FULL RANGE' wouldn't be a loop at all, it could just go straight to the mixer-section. He could put the full range-effect before or after this box so I don't think there would need to be a loop for the full range.

I need a vero-layout 'cos I can't do anything with just a schematic. So sorry about that. :)

I don't need them to be on one veroboard. Could be 3 different boards.

I've sourced the splitting: https://tagboardeffects.blogspot.fi/2014/08/amz-2-channel-splitter.html
and the mixing: https://tagboardeffects.blogspot.fi/2013/01/2-channel-mixer.html
but I can't find a layout for the crossover. :/

Oh, and the Phase-reverse switch would be a welcome addon too. Quess that could be just a SPDT-switch with a simple buffer-circuit on the other lug?


Hope you can help me out. :)

Cheers,
Riku

EBK

I'll drop this link here for now in case someone can use it to help you.  http://sound.whsites.net/project148.htm
  • SUPPORTER
Technical difficulties.  Please stand by.

samhay

#2
You don't want a crossover, you want a high pass filter.
This can be as simply as a series capacitor followed by a (variable) resistor to ground, but if you are not willing to look at a schematic, then there isn't much more I can say or do.
I'm a refugee of the great dropbox purge of '17.
Project details (schematics, layouts, etc) are slowly being added here: http://samdump.wordpress.com

Kummeli

Quote from: samhay on December 13, 2016, 02:39:48 AM
You don't want a crossover, you want a high pass filter.
This can be as simply as a series capacitor followed by a (variable) resistor to ground, but if you are not willing to look at a schematic, then there isn't much more I can say or do.

Ah ok, can the high pass filter get good attenuation? Say -18db or so?
I mean I can read a schematic but I don't know how to convert it to a verolayout. :)

samhay

>good attenuation

Are you are referring to how steep the filter is? If so, this is determined by how many 'poles' the filter uses:


Do you want the cutoff/corner frequency to be variable?
It easy to make a filter which has variable corner frequency, and is reasonably easy to make a steep filter. It is not particularly easy to make a steep filter with variable corner frequency.

I'm a refugee of the great dropbox purge of '17.
Project details (schematics, layouts, etc) are slowly being added here: http://samdump.wordpress.com

Kummeli

Quote from: samhay on December 13, 2016, 10:31:43 AM
>good attenuation

Are you are referring to how steep the filter is? If so, this is determined by how many 'poles' the filter uses:


Do you want the cutoff/corner frequency to be variable?
It easy to make a filter which has variable corner frequency, and is reasonably easy to make a steep filter. It is not particularly easy to make a steep filter with variable corner frequency.

Yeah steep was the word. :)
And the slope would "have to" be atleast -18dB/octave to be efficient enough for this purpose IMO.

The cutoff frequency would be better if it was variable via a potentiometer, but maybe a fixed frequency could work too. Say 400Hz or so.

I was thinkin.. Would it be better to have the low and high loops and blend them in the mixing-section rather than the 'full range' and high loops I was after?
Could the blending be easier that way? Propably not so much possible phase issues? Just guessing though..

samhay

18 dB/octave is a 3-pole filter. If that's really what you want/need, then making it variable is quite difficult.
If you want low-pass too, then that's quite possible, but uses about twice as many parts.
I'm a refugee of the great dropbox purge of '17.
Project details (schematics, layouts, etc) are slowly being added here: http://samdump.wordpress.com

robthequiet

Just to help the design theory, would it be possible to find out which effects are being used on the high side? I'm wondering if the solution might be a simple splitter, then have an equalizer dedicated to the effected side, then mixing them in an A/B/Y pedal -- might give better control over the tone, and be more flexible than putting the signal and frequency split into the same box. Oh, and maybe this would help, http://runoffgroove.com/splitter-blend.html

Kummeli

As of now there would be a Bass Muff, Boss Bass OD on the high-loop. Maybe a chorus and a bass synt too.
On the 'full range' or low loop there would be an octaver and a bass whammy.

The idea is to keep the bottom end clean and clear.

Searched the forum a bit and stumbled across this: http://www.diystompboxes.com/smfforum/index.php?topic=112437.msg1040196#msg1040196

Would that work? As a splitter and crossover.

samhay

The circuit that that layout is probably for  - reading back through the thread; I'm not going to trace it to find out - seems to use a 1-pole (6 dB/octave) high pass filter and then derive the complementary low pass filter from there.
The circuit topology is sound, but it won't give you a steep filter. However, you could switch between low pass and all-pass output using a SPST switch, which is maybe a nice feature.
I'm a refugee of the great dropbox purge of '17.
Project details (schematics, layouts, etc) are slowly being added here: http://samdump.wordpress.com

robthequiet

#10
I think the original post is the best way, i.e. full range pass-through and a high-pass for the bright side. Only need to compute one corner frequency to keep the mud out of the fuzz side. The idea of a crossover may be overthinking.

Now, making something variable, you may be getting into a semi-parametric eq territory if you want to vary the frequency, but a non-adjustable Q and allowing the mixer to control level saves a couple of pots. A circuit over on Rod Elliott's site might be a choice, http://sound.whsites.net/project155.htm, depending on how complex you would like it. A variable cutoff means less tinkering in the design as it can be adjusted in session.

edit: Why not go ahead and try the AMZ splitter, just to get the separation done. Then you need only build a filter to insert in the chain, as a separate project to keep things flexible.

Kummeli

Ok, I think I'll settle for something less than -18dB.. -12db will do.

So, I tried to draw my first ever schematic and verolayouts..



Please be gentle 'cos I've never done this before. :)

I used the suplied links as a reference. With this I "should" get a -12dB cut at variable frequency of about 80Hz to 420Hz?

Or am I completely lost here? haha. :)


robthequiet

#12
Just a couple of things -- first off, I would definitely breadboard this project first to avoid having modifications on strip board. I also see that you might want your input resistor to be more like 1M -- 22K to ground will load down the signal. You might also try a .01 uF capacitor at the input. Your variable resistors are meant to be a ganged pot, I believe.

So, put it on the breadboard and see what you have, and post back.

PRR

The audio part appears to be for a BiPolar supply, but I only see single supply on the layout.
  • SUPPORTER