Passive Filters

[YouAre]

I'm trying to get an understanding of passive filters, but i'm having a lot of trouble finding information on the internet. My main goal is to have unique tone controls for my guitar (les paul).

Its my understanding that most guitars have a Hi-cut filter. The hot signal is split into a signal to the output, and a signal to the tone pot, which dumps the highs that pass through the capacitor to the ground.

Can any one direct me to some layouts/schematics of different simple tonal filters which could include:
low pass/low cut filters
high pass/high cut filter
mid pass/mid cut filter ??

Thanks,
Murad

[phaeton]

http://www.generalguitargadgets.com/richardo/distortion/index.html


That right there is about the best start anyone could offer, IMHO.  It's a little ways down.



I wish I could meet this Richardo guy and shake his hand. 

[YouAre]

thanks! really interesting read!

Does anyone know how to use an inductor to cut mids? i've read up on this a little bit before. Using a cap as a hi-pass filter and an inductor as a low pass, you could cut the mids? Anyone have any experience with this?

[R.G.]

A capacitor has an impedance that decreases with increasing frequency. An inductor has an impedance that increases with increasing frequency.

When you put two of them together, you can place them in series or parallel.

In series, there is a "hole in the middle" where both are at their lowest relative impedance. The cap impedes low frequencies, the indutor high freuquencies, and only where they have equal impedances does much get through. This is at the frequency where Xc = Xl, or F = 1/(2*pi*SQRT(L*C)).

In parallel, the cap shunts highs through, the inductors shunts lows through, so there is a "post in the middle" where the combination is high impedance. Otherwise they pass signal. Same freuqency for the same reasons.

So if you want to block mids, you can use a parallel L-C pair in series with your signal. Or you can use a series L-C to ground to short out the mids.

The width of the notch depends on how much resistance is associated with the L-C parts. With zero resistance, the notch or post is infinitely thin. With real components the bandwidth can be quite selective. By adding more resistance you can widen it out to anything you like.

Note that a twin T filter made of only Rs and Cs will also notch out mids. It's the only R-C filter that can have an infinitely deep notch if you tune it correctly. This might be a good place to start for a guitar mids filter.

[YouAre]

A capacitor has an impedance that decreases with increasing frequency. An inductor has an impedance that increases with increasing frequency.

When you put two of them together, you can place them in series or parallel.

In series, there is a "hole in the middle" where both are at their lowest relative impedance. The cap impedes low frequencies, the indutor high freuquencies, and only where they have equal impedances does much get through. This is at the frequency where Xc = Xl, or F = 1/(2*pi*SQRT(L*C)).

In parallel, the cap shunts highs through, the inductors shunts lows through, so there is a "post in the middle" where the combination is high impedance. Otherwise they pass signal. Same freuqency for the same reasons.

So if you want to block mids, you can use a parallel L-C pair in series with your signal. Or you can use a series L-C to ground to short out the mids.

The width of the notch depends on how much resistance is associated with the L-C parts. With zero resistance, the notch or post is infinitely thin. With real components the bandwidth can be quite selective. By adding more resistance you can widen it out to anything you like.

Note that a twin T filter made of only Rs and Cs will also notch out mids. It's the only R-C filter that can have an infinitely deep notch if you tune it correctly. This might be a good place to start for a guitar mids filter.

wow thanks!
I looked up the Twin T filter, and its got a lot more parts than an LC filter which would require only a cap and an inductor. I actually want to put this as a tone control in my guitar, so if i had a Series LC to ground, would that be layed out the same as the tone control which has the capacitor that goes to ground, or would it be a totally different lay out.

And for that equation you gave me, what are the conversion factors for Henries and Farads? I would mostly be looking to get things in terms of mH and uF.

Thanks,
Murad

[d95err]

I looked up the Twin T filter, and its got a lot more parts than an LC filter which would require only a cap and an inductor. I actually want to put this as a tone control in my guitar, so if i had a Series LC to ground, would that be layed out the same as the tone control which has the capacitor that goes to ground, or would it be a totally different lay out.

And for that equation you gave me, what are the conversion factors for Henries and Farads? I would mostly be looking to get things in terms of mH and uF.

Checkout the "Varitone". That's the 6-position switch passive switch tone control found in some Gibson guitars. It's an LC notch filter that cuts mids in various frequencies. The inductor is fairly big, about 7 Henries.

http://www.blueshawk.info/official_gibson_schematic.htm

[R.G.]

if i had a Series LC to ground, would that be layed out the same as the tone control which has the capacitor that goes to ground, or would it be a totally different lay out.

A series LC to ground would act like mid scoop at the LC frequency, so yes, it acts like the treble cut control. In fact, you may be able to use the cap that's already in there.

And for that equation you gave me, what are the conversion factors for Henries and Farads? I would mostly be looking to get things in terms of mH and uF.

"milli" = m = 1/1000, so 1mH = 1/1000th H
"micro" = u = 1/ 1,000,000, so 1uF = 1/1,000,000 F.

Just for grins, "kilo" = k = x 1000, so 1k ohm is 1000 ohms; "mega" = M = x1,000,000, so 1M = 1,000,000 ohms.

"Nano" = n = 1/1,000,000,000, so 1nF = 1/1000th of 1uF, or 0.001uF.
"pico" = p = 1/ 1,000,000,000,000 so 1pF = 1/1000th of a nF, or 0.001nF, or 0.000000001uF or  0.000000000001F

[YouAre]

if i had a Series LC to ground, would that be layed out the same as the tone control which has the capacitor that goes to ground, or would it be a totally different lay out.

A series LC to ground would act like mid scoop at the LC frequency, so yes, it acts like the treble cut control. In fact, you may be able to use the cap that's already in there.

And for that equation you gave me, what are the conversion factors for Henries and Farads? I would mostly be looking to get things in terms of mH and uF.

"milli" = m = 1/1000, so 1mH = 1/1000th H
"micro" = u = 1/ 1,000,000, so 1uF = 1/1,000,000 F.

SO i did some calculations based on this equation. Using a 500mH Inductor out of my Vox wah (sneaky sneaky!), so that's .5 henries And I figured, i could calculate with the cap thats in my guitar, so .047uf (divided by 1000000) and i got like 1000hz or something (dont have the paper with me, but its in that general area) Are my calculations correct? Can i use a cap (that you'd normally use in a guitar) and a wah inductor?? I guess it seems right cause these are values you'd find in a wah (which boosts mids considerably).

[darron]

if you want something interesting, i would definitely try the big muff tone control. at one point it takes out the bass, and the other end it takes out the treble, and there's a bit of a mid dip sound in the middle.

it's really easy to make, and entirely passive... if you scroll down there's a schematic here:
http://www.generalguitargadgets.com/richardo/distortion/index.html

[YouAre]

if you want something interesting, i would definitely try the big muff tone control. at one point it takes out the bass, and the other end it takes out the treble, and there's a bit of a mid dip sound in the middle.

it's really easy to make, and entirely passive... if you scroll down there's a schematic here:
http://www.generalguitargadgets.com/richardo/distortion/index.html

yeah i checked out what it does in the duncan tonestack calculator (useful stuff!). I may use it, but right now i'm looking for something that cuts mids. But thanks for your suggestion. I may actualyl use it for the neck pickup. Could be interesting.

[YouAre]

Where's RG!? i know he can answers my last question!!

[Meanderthal]

 Craig Anderton's passive tone control from Electronic Projects for Musicians will do what you want. It's got a 5 way rotary switch to tune between the notch frequency/ mid scoop caps, and has a coil tap switch for the transformer to halve the frequency.  Very simple circuit, very similar to the varitone, can be put in a guitar, and uses a 42tm019 mouser transformer instead of an expensive wah inductor.

The big muff tonestack will cause a lot of volume loss...

[YouAre]

Craig Anderton's passive tone control from Electronic Projects for Musicians will do what you want. It's got a 5 way rotary switch to tune between the notch frequency/ mid scoop caps, and has a coil tap switch for the transformer to halve the frequency.  Very simple circuit, very similar to the varitone, can be put in a guitar, and uses a 42tm019 mouser transformer instead of an expensive wah inductor.

The big muff tonestack will cause a lot of volume loss...

Is this Project Linked online?

[Meanderthal]

 Give me about an hour... I'll draw a schematic.

[YouAre]

Give me about an hour... I'll draw a schematic.

Thanks a lot!!

[Meanderthal]

 Ok, here 'tis...

[YouAre]

Ok, here 'tis...

Ok so by looking at this, i'm gonna guess that one of the middle values will be the mids notch? What do you think would happen if i used a 500k pot for depth (the value of my guitar pot) I see that i can definetely just pick one of the Caps if i didn't need the option to choose between all the caps. And thanks again!

[Meanderthal]

 Here's the frequencies (hz)given for the notches:

Switch position

s1      s2 half coil    s2 full coil

1         1015              540
2           755             380
3           560             240
4           430             170
5           260             107

Most of these are what could be called mids.

I think a 500k pot would be ok.

I'm not 100% sure that 47k resistor is necessary either, maybe could be lowered or eliminated because of the guitar's volume pot already being there. Being passive there will be some signal loss, but it should be acceptable.

[YouAre]

Awesome! thanks for doing the calculations. I did some reverse calculations, and apparently the inductor is about 2.5 henries. Am i right? I used .01 uF for C, 1015 for F, and basically solved for L. (this was good practice for the calc exam I have on monday)

[Meanderthal]

 LOL! Yeah, it's 2.5 to 3 henries per half of the transformer coil. And, to give proper credit, I didn't do any calculations, Craig Anderton did way back in the 70s when he wrote the book I got this out of. All that talk about L-C passive filters rang a bell, and I dug it out...