Mid notch control

[bendo031]

Hi all.

I'm new to this forum and this is my first post.

I recently undertook to install a notch filter in my 18watt EF86 amp. I found a schematic online somewhere for a variable 400hz notch filter, which works great. I wired it with a pot going to ground, so the intensity of the scoop could be varied.

Im not an experienced builder or electronic engineer, so my technical knowledge is somewhat limited. I would like some explaination of the circuit, so as to give me some insite of how to calculate the frequencies etc. I think the circuit is a "bridged T" type.

Any advice?

Thanks in advance.

[martin manning]

This circuit is similar to a bridged attenuator (as referenced in the diagram) but in that case the pot wiper is grounded and the "bottom" pin goes to the C3-C1 node. You can find references to that circuit on the web.

To calculate the center frequency I'd select the cap value first, say 0.1uF (100nF) and then find R for the frequency you want as R = 159155/(fC), where C is in nF and R is in k-ohm. For 400Hz, that's 3.98k, so use the closest standard, 3.9k. Then, the pot is supposed to be 6*R, so thst would be 25k. Checking those values with a simulation, I get a -2dB scoop centered at 400Hz at the mid range of a linear pot, but it moves up to about 700Hz at the max setting, where the scoop is -6dB. You might want to design it for ~200Hz (200nF cap and 3k3 resistor) if you want the full-on scoop at 400Hz. It's also about a decade wide (factor of ten in frequency), so it will have some effect over a wide range of frequencies.

You have already wired it up and tried it? What values did you use?

[Reeltarded]

I am TOTALLY interested in working out a 5-6 position notch. This is very interesting.

[martin manning]

The catch is you have to vary the cap value (all three) to vary the frequency so you'd need a lot of caps and a four-pole switch.

[LeftyStrat]

My favorite notch filter is the Craig Anderton Passive Tone control.

[Reeltarded]

Aha.

Thanks for that one, Lefty! I go think now.

[bendo031]

This circuit is similar to a bridged attenuator (as referenced in the diagram) but in that case the pot wiper is grounded and the "bottom" pin goes to the C3-C1 node. You can find references to that circuit on the web.

To calculate the center frequency I'd select the cap value first, say 0.1uF (100nF) and then find R for the frequency you want as R = 159155/(fC), where C is in nF and R is in k-ohm. For 400Hz, that's 3.98k, so use the closest standard, 3.9k. Then, the pot is supposed to be 6*R, so thst would be 25k. Checking those values with a simulation, I get a -2dB scoop centered at 400Hz at the mid range of a linear pot, but it moves up to about 700Hz at the max setting, where the scoop is -6dB. You might want to design it for ~200Hz (200nF cap and 3k3 resistor) if you want the full-on scoop at 400Hz. It's also about a decade wide (factor of ten in frequency), so it will have some effect over a wide range of frequencies.

You have already wired it up and tried it? What values did you use?

Thanks for the response Martin.

I used this schematic, which I found randomly online somewhere. For some reason I couldn't find it when I made my first post, and posted what I thought was close to what I'd used. After looking again, I realize they are completely different.

I originally tried a DR.Z. type one knob control (ala Stang-Ray or Ghia) but I didn't like the way it sounded in this circuit, so looked for something simpler.

After installing this, I liked the result. Seems to go from almost Blackface Fender type mid scoop, and gain loss, all the way to voxy mid bark, at full rotation.

Somehow I think it could still be improved for my application, as good as it is. I'd like to understand what value(s) to alter, so as to tune it exactly as I'd like.

Thanks for the help.

[Reeltarded]

That looks almost too easy. I need a couple parts.

[wicker]

Here is simulation of filter posted by bendo, what I dont like about it, is that it's not very steep, you might try twin T filter. Values for R3 in Twin T are from 1k to 100k.

[VacuumVoodoo]

grr..double...

[VacuumVoodoo]

A well over 1 octave shift in mid dip center frequency (it's too wide to be called a notch) over it's level adjustment range is too much to be really useful. Unless you call it an effect and sweep it with a foot pedal....
There's a way to use twin-T that has a much smaller shift while allowing at least 20db adjustment range.
Later tonight, must get out and do my 10.000 steps power walk. Doctors orders.
Ta'ta.

[tubeswell]

You listen to that doctor.

[rp]

Later tonight, must get out and do my 10.000 steps power walk. Doctors orders. Ta'ta.

So how far from home does 5000 steps get you? Just curious.

[Reeltarded]

He's in Sweden.. to the border.

[VacuumVoodoo]

So how far from home does 5000 steps get you? Just curious.

Depends on how long steps I take, but somewhere between 3 and 4 miles.

He's in Sweden.. to the border.

Well, maybe not the border but certainly to the waterfront.

Anyway, here's the circuit and frequency responses.
Driver stage parameter influence low end and flatness of neutral response.
Just for fun the center frequency is 440Hz
R4 sets max depth of the dip.
Change the frequency by changing R1,R2 & R3, keep R1=R2 and R1/R3=3
Textbook relationship R1/R3=2 doesn't sound good.
The formula for different R values for other frequencies is....obvious
Have fun!