synthy filters for guitar..

[Strategy]

Interesting, clip sounds almost phaser like!

[nocentelli]

Could anyone recommend an envelope detector-> CV output circuit that might well with guitar and might be suitable for these synth-y filters? I've tried the Meatball arrangement, but it's a bit tricky to get it too work, and the "attack" and "decay" controls are so interactive. I only end up leaving one at max and one at minimum, so they are somewhat superfluous.

[garcho]

Well, it is an Escobedo-related thread, check out the Phuncgnosis?

Read this thread.

[garcho]

From Mr. Escobedo:

The CV signal is injected into the signal path as a kind of fluctuating DC component, causing the LED(s) to vary in resistance, which dynamically tunes the filter.

[Mark Hammer]

Some 20 years ago, maybe more, I built an Anderton/EPFM Super Tone Control into a wah shell, sweeping it with a dual-ganged pot, and installing a reverse-sweep switch to flip the outside lugs of the dual pot around so that I could change the direction of sweep for toe-forward.  The STC is a basic state-variable filter, as found in the Mutron, Meatball, and a zillion other things.

The unit would provide lowpass, bandpass, highpass, and band-reject (notch) if you combined lowpass and highpass.  You wouldn't think that a foot-operated pedal requires a direction-reverse, the way that something like an autowah does.  However, where sweeping toe-down moves the centre frequency of a bandpass and band-reject upward, and the corner frequency of a lowpass upwards, it also completely guts the outcome of a highpass filter as you sweep forward.   One's natural inclination for a foot-swept pedal is that you lean forward for greater emphasis, so getting a wimpier sound in toe-down just works counter to your inclinations.  So I built in a means to reverse the direction of sweep. Worked great. 

Indeed, modding any autowah that uses a state-variable filter to be controlled by a foot-pedal will get you some nifty, and synth-ey, sounds.  Part of that is because the sweeps you can get with foot-control start to approximate what you can get with long attack times...but without requiring a trigger and transient generator.  The reverse-sweep also provides for the sorts of sweeps that are very hard to get from an autowah.

Just a caveat that not all dual-ganged pots will fit in the available space in wah shells, or accept the gear in the rack-and-pinion systems.  I suppose one could use an existing single pot to control the current feeding one or two LEDs that govern a pair of LDRs.

[pappasmurfsharem]

Quite frankly, just about ANY filter sounds more "synth-ey" on a guitar if it has a relatively quick decay.  I modded up a Bass Balls with a variable decay, filter stagger, and filter blend control, that you can hear here.  You'll have to excuse the overly lengthy techtalk at the beginning, and the fluorescent light him picked up on the audio, but you can still hear what that various controls are able to do to an otherwise nondescript circuit.  For my part, I think it nails some of the sounds of EHX's old Mini-synth keyboard.

Phunkpunk: http://www.youtube.com/watch?v=eoAIKHFGOcQ
Minisynth: http://www.youtube.com/watch?v=Mi9Kh4bgl5Y

And one should also note that audible envelope ripple can be "cured" in one of two ways: making the decay time slower so that ripple is more averaged out, or making it faster, so that ripple doesn't have a chance to be heard.  So not only does a faster decay sound more synthey, but it also has less ripple.  Win-win.

Hey Mark,

On your Phunkpunk for your decay knob I believe your video states that you  replaced the 330k with a 47k and a 500kto1M pot. I assume this controll responds in a linear fashion? So a linear pot would be most beneficial?

[Jdansti]

https://dl.dropboxusercontent.com/u/7464107/2q%26dfilters.mp3

q&d experiments clip

1st hi filter...

2nd low filter

3rd both filters..

Sounds great Rob! At first I thought that you were playing a single note and the circuit was playing the 16th notes while it was sweeping, but I'm sure you were alternate picking the 16th notes. Very promising!  Time for some "Teenage Wasteland"!

[deadastronaut]

^

ok ive been breading an lfo here..(ricks)
http://www.diystompboxes.com/smfforum/index.php?topic=74294.msg623245#msg623245

and hooking it up to the CV...( without led) but i'm not getting the Q&D filters full sweep....just a little really..its working, but much weaker than i thought it would.

should i get rid of the freq pot entirely?..

should i be using another lfo to drive the filter sweep?

[Mark Hammer]

Quite frankly, just about ANY filter sounds more "synth-ey" on a guitar if it has a relatively quick decay.  I modded up a Bass Balls with a variable decay, filter stagger, and filter blend control, that you can hear here.  You'll have to excuse the overly lengthy techtalk at the beginning, and the fluorescent light him picked up on the audio, but you can still hear what that various controls are able to do to an otherwise nondescript circuit.  For my part, I think it nails some of the sounds of EHX's old Mini-synth keyboard.

Phunkpunk: http://www.youtube.com/watch?v=eoAIKHFGOcQ
Minisynth: http://www.youtube.com/watch?v=Mi9Kh4bgl5Y

And one should also note that audible envelope ripple can be "cured" in one of two ways: making the decay time slower so that ripple is more averaged out, or making it faster, so that ripple doesn't have a chance to be heard.  So not only does a faster decay sound more synthey, but it also has less ripple.  Win-win.

Hey Mark,

On your Phunkpunk for your decay knob I believe your video states that you  replaced the 330k with a 47k and a 500kto1M pot. I assume this controll responds in a linear fashion? So a linear pot would be most beneficial?

I wish I could tell you, but a guy saw the video, contacted me, and bought it earlier this year.  But, given how the ear/brain works, I suspect that log or reverse-log would be optimal; log if the pot is wired to get you smaller resistances (i.e., faster decay) going clockwise, and reverse-log if clockwise gets you slower decay.  Alternatively, there is absolutely nothing wrong with using a 3-position toggle to get fast/medium/slow decay (47k, 330k, 1M).

[garcho]

should i get rid of the freq pot entirely?

try lowering the 100k CV in resistor?

are you including the 100k on the LFO output?

[deadastronaut]

hi gary,

cv: tried lowering , and totally without the  100k..

got rid of lfo out 100k..

i got a more defined sweep...but its still not the full sweep that the freq pot range is..

[nick d]

           A 100k pot across a 9v supply will sweep the CV from 0 to 9V , an LFO running from a 9V supply will have an output swing of quite a bit less , perhaps 2/3 . So , the overall effect will be milder . ( I think ! )

[deadastronaut]

yeah makes sense i guess..

so i need an lfo that sweeps 0-9v...

any ideas?..links?..

just found this from paul

http://www.diystompboxes.com/smfforum/index.php?topic=87861.msg739719#msg739719


edt:  just tried that,^  i'm getting 2.7 to 7.3 voltage swing...

so am still missing the crucial low and top end...hmmmm..

[samhay]

Adding gain to your second op-amp will help, but unless you use a rail-to-rail op-amp, the best you can hope for is about 2-7 V. If you have an LM358, that can pull down to 0, which might be helpful.

I had previously sim'd the filter, and just took another look. You're using something like this right?
http://folkurban.com/Site/QDVCF-714.html
As you have probably discovered, the sweep works all the way from 0 to 9V. Bass at 0V, trebble peak at 9V. If you are stuck with ca. 2-7 V, then you probably want to shift the filter response 'down', so you still get a nice resonant peak at 7V. If you can't get there with by reducing the value of the 'resonance' resistor, then you might want to try increasing the value of the two series resistors a little.

[deadastronaut]

cheers  sam. yep that filter...

pretty sure i have a few 358's...

is it a case of just swapping the ic, or will (as i suspect) need a different lfo then.?...

[samhay]

Not sure that I have tried, but I can't see why that LFO wouldn't work with a 358. You might want to play with the values of the 100k voltage divider resistors (to the (+) input) to centre the LFO between the rails - say at about 3.5 - 4.0 V.

[Lurco]

hi gary,

cv: tried lowering , and totally without the  100k..

got rid of lfo out 100k..

i got a more defined sweep...but its still not the full sweep that the freq pot range is..

You can`t just lower the CV 100k, or else the lowest possible frequency will go upward. Fmin is determined by the resistance between the two .01µf caps and (virtual) ground. This resistance is the parallel combination of the CV 100k, the Freq. 100k (plus the parallel resistances of the divided pot-halves), and the dynamic resistance of the LED. The latter is controlled by the current running through the LED. If you want to increase the LED current for accepting a low-voltage LFO swing by decreasing the LED series resistor (i.e.: the CV 100k) to e.g. 33k, you might want to adapt the whole twin-t filter. This can be done by multiplying the cap values by 3, and dividing the resistor values by 3. In theory ar least. (this concerns the left schematic here: http://folkurban.com/Site/QDVCF-714.html).

[deadastronaut]

@sam:  tried the 358. much better for the lfo  sweep... 

[samhay]

That's the kind of fix we like.

[garcho]

Thanks for explaining Lurco!