Let me first put my question into context. I was looking at the schematic for the small clone, and thought that you may be able to come up with a combo flanger/chorus out of this design.
As I understand the differences between chorus and flanger, the flanger has a shorter delay, feedback of the delay signal, and where the chorus modulation is typically a sine wave, the flanger typically uses a triangle wave.
When I breadboarded the lfo of the small clone, I saw a triangle wave output, so I was thinking, why not add a feedback path, and twiddle with the clock to get a simple switchable chorus/ flanger?
My first instinct is to try to graft Scott Swartz PT-80 compandor circuit to the small clone. I figure feedback is likely to deteriorate noise performance. I have yet to try this, but Scotts compandor design looks like it could be lifted directly out of the PT-80. That would give better noise and a feedback path. Any thoughts on this idea? Does anyone see any mods that would need to be made to the compandor?
All this is moot if the lfo and clock cannot be adjusted to get into the flanging range. It seams that messing with the cap on pin 2 of the CD4047 should be able to slow down the clock to the chip.
To get to my main question, I was trying to get a handle on how the lfo modulates the clock that drives the delay chip for the small clone.
Basically, the lfo connects to the CD4047 chip by coupling the output of the depth control through a series 39k resistor and a 10uf cap in parallel. The 39k resistor couples to pin 1 through a series connection with a capacitor, pin 2 through a series connection with a diode, and pin 3 directly.
Now when I look at the CD4047 data sheet, the pin 2 is labeled a resistor, and pin 3 is called an RC common.
So what does the diode do? What is the resistance that the CD4047 "sees"?
How does the modulation of the lfo affect the resistance?
Any help would be greatly appreciated.
Assuming that the Small Clone produces acceptable noise performance at the moment, I don't think a compandor will add dramatically to flanger performance. Certainly the increase in HF clock rate needed for flanging puts the clock signal well out of the audio range so the primary source of noise becomes whatever hiss and grit is accumulated under normal functioning.
The range of the HF clock is shifted by the 150pf cap. This is analogous to the small value cap traditionally found right beside the MN3101, which is, in turn, traditionally found immediately adjacent to the MN3007 in other commercial flanger/chorus units. 'ake it bigger to increase the delay time and "thicken" the chorus tone. Make it smaller to shorten the delay and move it more into the flanging range.
Of course, where very slow sweep rates are useless on chorus units, they are pleasing on flangers, so you want to be able to shift the LFO frequency range as well. The 2.2 tantalum cap in the schematic/drawing for the Tonepad "Heladito" is the one to change. Increasing its value will slow down the LFO. The simplest way to do that is to tack on a parallel 10uf or 4.7uf cap for a suitable range shift. 4.7uf will give you a combined capacitance of 6.9uf, which will drop the slowest sweep rate by a little over an octave and a half. So if your slowest current sweep is 1hz, then it will become about .3hz or so, or one sweep every 3 seconds. 10uf will drop the LFO floor frequency even more.
Feedback would normally be achieved by tapping the delay signal at the junction of the 1uf cap, 220k and 20k resistors just prior to the wet/dry mixing point, and feeding it back to pin 3 of the first op-amp at the input through a DC-blocking cap and a resistor. I'll fudge and suggest a 1k resistor in series with a 50k pot. A 0.1uf cap ought to be fine.
Since feedback can run away on one, it is customary to include some peak-limiting diodes in the feedback path of wherever the mixing is done. Indeed, you'll see that in the schem for the BOSS BF-2. A pair of normal silicon switching diodes (1N914/4148) should be fine. The idea is not to generate clipping necessarily, but to make sure that the feedback loop does not result in ridiculous input signals to the BBD and oscillation. I suppose one could play with the diode aspect a bit, perhaps by sticking a series resistor in to make any clipping a little softer, or by selecting the diode type based on headroom concerns and effective signal measures at the input to the BBD itself. Remember that the 3-pole lowpass filter formed around Q1 can eat up some signal amplitude, and the input stage around IC1a adds some gain. Between all these various aspects, it may be that the ceiling needs to be set a little higher, OR a little lower, than the 500mv or so that a back-to-back pair of diodes will set.
Since the same stage is handling input mixing and feedback duties, whatever feedback limiting you build into IC1a will also have an impact on the level of the dry signal at pin 1 of IC1a. For that reason, you may find you need to tinker with the 22k resistor between IC1a and IC1b, perhaps by changing it to 20k, 18k or maybe even less, depending on what you hear.
That won't get you a stellar flanger, but it will deliver far more sounds than you get out of it presently.
Mark:
Your help is most appreciated. My goal is to have a multifunction device for those "lite" gigs where I don't want to lug a big pedal board around, but want to get as much bang for the buck out of my effects as possible. Therefore the tradeoff in flanger quality for compacting two pedals into one is acceptable.
I will try out your suggestions and let you know how things turn out.
Thanks for the great advice.