A mod to the AMZ mod to the BMP tonestack?

[mordechai]

I'm making a BMP for a friend, and he has asked not only for the mid-control feature, but also a feature that could change the Q of the mid scoop from wider to narrower.  Actually, I'm making it so that it also gets a mid bump rather than just flat mids, so this would also be a Q control for the mid bump.  I've noodled with the Duncan Tone Stack calculator and see different ways to shift the mid range scoop/bump, but I'm not sure how to widen or narrow the Q.  Anybody have a suggestion on how I could do this?

[antonis]

I'm not sure if I got you right but BMP tonestack actually is a mix of HPF & LPF (each of them is considered as  1^st order filter..)

Such as, they don't have a center frequency, hence no Q..
(in case of 2^nd order they should have an individual Q (peak amplitude at resonant frequency..)

2^nd order Low-pass filter example:



In case of you refer on Jack's Dual tone http://www.muzique.com/lab/dtone.htm you'll have to seperately adjust Q's with also some kind of active device..

[Mark Hammer]

Certainly the portion of the spectrum where the scoop falls, and the width of the scoop, can be varied.  I have a couple of BMPs in which I've installed toggles to move the lowpass section higher and/or the highpass section lower.  Both are useful.  But as Antonis wisely notes, with passive 1st-order sections, there isn't much opportunity to modify Q, only scoop width as I've done.

On the other hand, Jack himself has posted about using 2-pole passive high and lowpass sections in the tonestack.  That would alter the selectivity of the scoop.  Moving the scoop around, or adjusting its width would require use of dual-ganged pots.  While that would be an interesting exercise, from an end-user's point of view it might be easier to just build different BMPs.

[mordechai]

Thanks to both of you, and Mark...the "width" control actually sounds like what I am after (if that means that it can control how many frequencies are getting cut or boosted).  Can you elaborate on how that works?

[Mark Hammer]

Just change the cap values in the high and lowpass sections.  A 3-position on-off-on SPDT toggle for the cap-to-ground in the lowpass section, and another for the series cap in the highpass section, would provide for a lot of possibilities.  One could also tinker with the resistance values in each of those sections instead, but I suspect that would monkey with the relative volume balance more than is convenient.  Personally, I find that changing the corner frequency via the caps provides for a more usable - if not continuously variable - change in tone.

For the lowpass section, the stock cap to ground is 10nf.  Make it 5n6 (.0056uf) to raise the corner frequency from 480hz to 860hz.  Use the toggle to add in 4n7 to restore stock rolloff or .01uf to move the rolloff down to 310hz.  For the highpass section, swap out the stock 3900pf for 3300pf to move the corner frequency up to 1460hz, and use the toggle to add in either 620pf for the stock rolloff or 1200pf for a rolloff starting around 1070hz.

These are just suggestions.  Feel free to go bananas or more conservative.  But clearly, a lowpass at 860 and highpass at 1070 provides very little midscoop, considering how much 1-pole slopes would overlap, while 310 and 1460 would provide for much open space between.

[antonis]

Feel free to go bananas

Now, after Mark's provocation..

For moving up/down scoop (w constant width), use a dual gang pot (50k say, with proper caps values resizing..)
For widening/narrowing scoop "bandwidth", use two individual pots (each one in place of respective filter resistor)
For both the above, mix them.. 

[mordechai]

Antonis...I think I understood Mark's suggestion above, and the second of your suggestions, but I'm having trouble envisioning how to implement the dual gang pot.  Can you explain a bit further?  It sounds like an interesting option but I am not sure I understand what it would control.

[Mark Hammer]

Feel free to go bananas

Now, after Mark's provocation..

For moving up/down scoop (w constant width), use a dual gang pot (50k say, with proper caps values resizing..)
For widening/narrowing scoop "bandwidth", use two individual pots (each one in place of respective filter resistor)
For both the above, mix them.. 

That CAN be done, and will certainly "work", but will require adjusting the Tone control every time you adjust the shift control, because the balance of low and high will be altered.

[Steben]

Having a large "Q" in a 1st order passive filter is hard.

[mordechai]

But it could be done if I incorporated an active control somewhere into the circuit, right? 

[Mark Hammer]

Essentially what you'd do is stick an active parametric section where the current tonestack is.  That would allow for boost/cut, centre frequency, and Q.  This, in turn, would theoretically allow for moving a scoop of varying width/selectivity around.

The stock BMP tonestack always lets at least some bottom and top through, and merely changes the balance/mix of them, the question becomes one of whether you'd want/need any additional tone-shaping beyond the parametric section.  And my guess is that you probably would, if only to remove "the fiizzies".