Mid Notch / Mid Bump Filter - Duncan Tone Stack

[Rayman]

I was playing with the Duncan Tone Stack calculator attempting to find a filter that will give a mid notch to a mid bump, but also raise and lower the bass and treble respectively to give an overall even output, all with one control knob.  I found the following to do exactly what I wanted and was very happy with the results.

TSC - E-Series
R1 - 22K (Treble on 10)
R2 - 1K
R3 - 22k
R4 - 1M (Bass on 10)
R5 - 1 ohm (don't want resistance here)
C1 - 22n
C2 - 10n
C3 - 100u (don't really want capacitance here)
C4 - 1p (don't want a signal here)

If you "manually" vary the R2 resistor between 1K and 0, yeah I know "0" - I'm getting to my question:) you will get a smooth transistion from a mid bump at around 900Hz to a mid notch at around 900Hz.  What is bothering me is how can I take R2 to "0" and still have a signal?  That creates a direct path from the input to ground.  Could there be a glitch in the TSC calcs?  I hope not, cause this does exactly what I'm looking for.  Maybe I'm missing something.

Any expert advice is appreciated.

[H.Manback]

If you really want to figure out what's up, do a google to find the calculations for all the parameters of this filter. I strongly suspect that this calculation is a bit fruity...

Also you could simulate it with pspice or something. I'll see what I can come up with.

[Arn C.]

Would a "zero" ohm resistor help in this case?

Arn C.

[H.Manback]

Looks like it is kind of fruity. I made it in LTSpice and it was completely wrong with your values. The standard values from the program did work the same, but my guess is that it is a course simulation that doesn't work for strange values anymore.

[stm]

I tried your values in the TSC and definitely there is something wrong in there, probably due to numerical problem with the computation, like adding a large number with a small number, dividing by zero, etc.

Another proof of this not working properly can be seen with your values if you turn down the TREBLE control below 5. You will see how the circuit tries to have positive dB gain (curve squasehed against the top of the graph) starting only from resistors and capacitors. Not possible.

For a tone control like what you are looking for, take a look here: http://www.muzique.com/lab/tone3.htm

The BODY control does what you want, especially the Second Version. You can leave the TONE control fixed with resistors if you just need a single-knob tone control.

Regards,

stm

[Rayman]

thank you for your help.  I did check out the AMZ presence control which is where I started from.  That control is a great idea.  It is defineatley the closest I've found to do what I want, the only problem is that it doesn't change the bass and treble with respect to the mids, it only boosts and cuts the mids.  Also it shifts the notch slightly.

Again, thanks for your help.  I need to keep experimenting.

Ray

[Rayman]

H.Manback

I took a look at the LTSpice and downloaded.  I may be in way over my head, but can you just insert symbols and create a circuit?  If so how would you put in a battery (9vdc)?

Ray

[H.Manback]

Disclaimer: It's probably not the best program, but it's good enough for me.

To add a battery you add a component (F2) and type in voltage. Then you modify that voltage source to be 9v dc (or a bit higher for fresh batteries, use imagination )

About adding the symbols, yes you just throw them in and connect them with wire. One note though, opamps are a bit of a weird story. You have to use the opamp (ideal opamp) and add a spice directive saying '.lib opamp.sub'. The LT opamps in the libraries didn't work for me.

[Rayman]

Thanks,

This may not be perfect but it is alot better than trying to manipulate existing setups.  Thanks again.